1 Change log file for Exim from version 4.21
2 -------------------------------------------
7 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
9 PP/02 Make -n do something, by making it not do something.
10 When combined with -bP, the name of an option is not output.
12 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
15 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
16 $sender_host_name and config options to manage this, and basic check
19 PP/05 DSCP support for outbound connections and control modifier for inbound.
21 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
22 (Only plugin which currently uses this is kerberos4, which nobody should
23 be using, but we should make it available and other future plugins might
24 conceivably use it, even though it would break NAT; stuff *should* be
25 using channel bindings instead).
27 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
28 name; added for Sendmail compatibility; requires admin caller.
29 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
30 just ignore it); requires trusted caller.
31 Also parse but ignore: -Ac -Am -X<logfile>
34 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
36 TL/02 Add +smtp_confirmation as a default logging option.
38 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
39 Patch by Magnus Holmgren from 2007-02-20.
41 TL/04 Bugzilla 1281 - Spec typo.
42 Bugzilla 1283 - Spec typo.
43 Bugzilla 1290 - Spec grammar fixes.
45 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
47 TL/06 Add Experimental DMARC support using libopendmarc libraries.
49 TL/07 Fix an out of order global option causing a segfault. Reported to dev
50 mailing list by by Dmitry Isaikin.
52 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
54 JH/02 Support "G" suffix to numbers in ${if comparisons.
56 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
58 NM/01 Bugzilla 1197 - Spec typo
59 Bugzilla 1196 - Spec examples corrections
61 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
63 PP/09 Add gnutls_enable_pkcs11 option.
65 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
66 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
68 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
69 "acl {{name}{arg}...}", and optional args on acl condition
72 JH/05 Permit multiple router/transport headers_add/remove lines.
74 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
76 JH/07 Avoid using a waiting database for a single-message-only transport.
77 Performance patch from Paul Fisher. Bugzilla 1262.
79 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
82 JH/09 Add $headers_added variable, with content from use of ACL modifier
83 add_header (but not yet added to the message). Bugzilla 199.
85 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
86 Pulled from Bugzilla 817 by Wolfgang Breyha.
88 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
90 (nb: this is the same fix as in Exim 4.80.1)
92 JH/11 Add A= logging on delivery lines, and a client_set_id option on
95 JH/12 Add optional authenticated_sender logging to A= and a log_selector
98 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
100 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
101 advertise SMTP AUTH mechanism to us, instead of a generic
102 protocol violation error. Also, make Exim more robust to bad
103 data from the Dovecot auth socket.
105 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
107 When a queue runner is handling a message, Exim first routes the
108 recipient addresses, during which it prunes them based on the retry
109 hints database. After that it attempts to deliver the message to
110 any remaining recipients. It then updates the hints database using
113 So if a recipient address works intermittently, it can get repeatedly
114 deferred at routing time. The retry hints record remains fresh so the
115 address never reaches the final cutoff time.
117 This is a fairly common occurrence when a user is bumping up against
118 their storage quota. Exim had some logic in its local delivery code
119 to deal with this. However it did not apply to per-recipient defers
120 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
122 This change adds a proper retry rule check during routing so that the
123 final cutoff time is checked against the message's age. We only do
124 this check if there is an address retry record and there is not a
125 domain retry record; this implies that previous attempts to handle
126 the address had the retry_use_local_parts option turned on. We use
127 this as an approximation for the destination being like a local
128 delivery, as in LMTP.
130 I suspect this new check makes the old local delivery cutoff check
131 redundant, but I have not verified this so I left the code in place.
133 TF/02 Correct gecos expansion when From: is a prefix of the username.
135 Test 0254 submits a message to Exim with the header
139 When I ran the test suite under the user fanf2, Exim expanded
140 the header to contain my full name, whereas it should have added
141 a Resent-Sender: header. It erroneously treats any prefix of the
142 username as equal to the username.
144 This change corrects that bug.
146 GF/01 DCC debug and logging tidyup
147 Error conditions log to paniclog rather than rejectlog.
148 Debug lines prefixed by "DCC: " to remove any ambiguity.
150 TF/03 Avoid unnecessary rebuilds of lookup-related code.
152 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
153 Bug spotted by Jeremy Harris; was flawed since initial commit.
154 Would have resulted in OCSP responses post-SNI triggering an Exim
155 NULL dereference and crash.
157 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
159 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
160 Bug detection, analysis and fix by Samuel Thibault.
161 Bugzilla 1331, Debian bug #698092.
163 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
165 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
166 Server implementation by Todd Lyons, client by JH.
167 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
168 config variable "prdr_enable" controls whether the server
169 advertises the facility. If the client requests PRDR a new
170 acl_data_smtp_prdr ACL is called once for each recipient, after
171 the body content is received and before the acl_smtp_data ACL.
172 The client is controlled by bolth of: a hosts_try_prdr option
173 on the smtp transport, and the server advertisement.
174 Default client logging of deliveries and rejections involving
175 PRDR are flagged with the string "PRDR".
177 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
178 fclose(). Diagnosis by Todd Lyons.
180 PP/17 Update configure.default to handle IPv6 localhost better.
181 Patch by Alain Williams (plus minor tweaks).
184 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
185 This is now consistent with GnuTLS, and is now documented: the
186 previous undocumented portable approach to treating the option as
187 unset was to force an expansion failure. That still works, and
188 an empty string is now equivalent.
190 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
191 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
192 not performing validation itself.
194 PP/20 Added force_command boolean option to pipe transport.
195 Patch from Nick Koston, of cPanel Inc.
197 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
200 TF/04 Added udpsend ACL modifer and hexquote expansion operator
202 PP/21 Fix eximon continuous updating with timestamped log-files.
203 Broken in a format-string cleanup in 4.80, missed when I repaired the
204 other false fix of the same issue.
205 Report and fix from Heiko Schlichting.
208 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
209 Report from Prashanth Katuri.
211 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
212 It's SecureTransport, so affects any MacOS clients which use the
213 system-integrated TLS libraries, including email clients.
215 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
216 using a MIME ACL for non-SMTP local injection.
217 Report and assistance in diagnosis by Warren Baker.
219 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
221 JH/16 Fix comparisons for 64b. Bugzilla 1385.
223 TL/09 Add expansion variable $authenticated_fail_id to keep track of
224 last id that failed so it may be referenced in subsequent ACL's.
226 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
229 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
230 ldap library initialization, allowing self-signed CA's to be
231 used. Also properly sets require_cert option later in code by
232 using NULL (global ldap config) instead of ldap handle (per
233 session). Bug diagnosis and testing by alxgomz.
235 TL/12 Enhanced documentation in the ratelimit.pl script provided in
236 the src/util/ subdirectory.
238 TL/13 Bug 1301 - Imported transport SQL logging patch from Axel Rau
239 renamed to Transport Post Delivery Action by Jeremy Harris, as
242 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
243 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
244 redis_servers = needs to be configured which will be used by the redis
245 lookup. Patch from Warren Baker, of The Packet Hub.
247 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
254 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
256 This, or similar/improved, will also be change PP/11 of 4.82.
262 PP/01 Handle short writes when writing local log-files.
263 In practice, only affects FreeBSD (8 onwards).
264 Bugzilla 1053, with thanks to Dmitry Isaikin.
266 NM/01 Bugzilla 949 - Documentation tweak
268 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
271 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
273 PP/02 Implemented gsasl authenticator.
275 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
277 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
278 `pkg-config foo` for cflags/libs.
280 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
281 with rest of GSASL and with heimdal_gssapi.
283 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
284 `pkg-config foo` for cflags/libs for the TLS implementation.
286 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
287 properties get this fed in as external SSF. A number of robustness
288 and debugging improvements to the cyrus_sasl authenticator.
290 PP/08 cyrus_sasl server now expands the server_realm option.
292 PP/09 Bugzilla 1214 - Log authentication information in reject log.
293 Patch by Jeremy Harris.
295 PP/10 Added dbmjz lookup type.
297 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
299 PP/12 MAIL args handles TAB as well as SP, for better interop with
300 non-compliant senders.
301 Analysis and variant patch by Todd Lyons.
303 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
304 Bug report from Lars Müller <lars@samba.org> (via SUSE),
305 Patch from Dirk Mueller <dmueller@suse.com>
307 PP/13 tls_peerdn now print-escaped for spool files.
308 Observed some $tls_peerdn in wild which contained \n, which resulted
309 in spool file corruption.
311 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
312 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
313 or write after TLS renegotiation, which otherwise led to messages
316 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
317 as a tracking header (ie: a signed header comes before the signature).
318 Patch from Wolfgang Breyha.
320 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
321 comma-sep list; embedded commas doubled.
323 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
325 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
327 Report and patch from Dmitry Banschikov.
329 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
330 Removed SSL_clear() after SSL_new() which led to protocol negotiation
331 failures. We appear to now support TLS1.1+ with Exim.
333 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
334 lets Exim select keys and certificates based upon TLS SNI from client.
335 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
336 before an outbound SMTP session. New log_selector, +tls_sni.
338 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
339 NULL dereference. Report and patch from Alun Jones.
341 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
342 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
343 Not seeing resolver debug output on NetBSD, but suspect this is a
344 resolver implementation change.
346 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
347 Left warnings. Added "eximon gdb" invocation mode.
349 PP/21 Defaulting "accept_8bitmime" to true, not false.
351 PP/22 Added -bw for inetd wait mode support.
353 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
354 locate the relevant includes and libraries. Made this the default.
356 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
357 Bugzilla 1246, report and most of solution from Tomasz Kusy.
359 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
360 This may cause build issues on older platforms.
362 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
363 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
364 gnutls_require_mac & gnutls_require_protocols (no longer supported).
365 Added SNI support via GnuTLS too.
366 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
368 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
370 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
371 Applied second patch from Janne, implementing suggestion to default
372 multiple-strings-in-record handling to match SPF spec.
374 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
376 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
377 read-only, out of scope).
378 Patch from Wolfgang Breyha, report from Stuart Northfield.
380 PP/29 Fix three issues highlighted by clang analyser static analysis.
381 Only crash-plausible issue would require the Cambridge-specific
382 iplookup router and a misconfiguration.
383 Report from Marcin Mirosław.
385 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
387 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
388 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
389 As part of this, removing so much warning spew let me fix some minor
390 real issues in debug logging.
392 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
393 assignment on my part. Fixed.
395 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
396 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
397 Janne Snabb (who went above and beyond: thank you).
399 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
400 string otherwise requires a connection and a bunch more work and it's
401 relatively easy to get wrong. Should also expose TLS library linkage
404 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
405 64-bit ${eval} (JH/03).
407 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
408 GNU libc to support some of the 64-bit stuff, should not lead to
409 conflicts. Defined before os.h is pulled in, so if a given platform
410 needs to override this, it can.
412 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
413 protection layer was required, which is not implemented.
414 Bugzilla 1254, patch from Wolfgang Breyha.
416 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
417 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
418 tls_dhparam take prime identifiers. Also unbreak combination of
419 OpenSSL+DH_params+TLSSNI.
421 PP/39 Disable SSLv2 by default in OpenSSL support.
427 PP/01 Solaris build fix for Oracle's LDAP libraries.
428 Bugzilla 1109, patch from Stephen Usher.
430 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
432 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
435 TF/02 Fix a couple more cases where we did not log the error message
436 when unlink() failed. See also change 4.74-TF/03.
438 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
439 lock up or crash if it happened to be inside a call to libc when it
440 got a SIGUSR1 from exiwhat.
442 The SIGUSR1 handler appends the current process status to the process
443 log which is later printed by exiwhat. It used to use the general
444 purpose logging code to do this, but several functions it calls are
445 not safe for signals.
447 The new output code in the SIGUSR1 handler is specific to the process
448 log, and simple enough that it's easy to inspect for signal safety.
449 Removing some special cases also simplifies the general logging code.
450 Removing the spurious timestamps from the process log simplifies
453 TF/04 Improved ratelimit ACL condition.
455 The /noupdate option has been deprecated in favour of /readonly which
456 has clearer semantics. The /leaky, /strict, and /readonly update modes
457 are mutually exclusive. The update mode is no longer included in the
458 database key; it just determines when the database is updated. (This
459 means that when you upgrde Exim will forget old rate measurements.)
461 Exim now checks that the per_* options are used with an update mode that
462 makes sense for the current ACL. For example, when Exim is processing a
463 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
464 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
465 must specify per_mail/readonly. If you omit the update mode it defaults to
466 /leaky where that makes sense (as before) or /readonly where required.
468 The /noupdate option is now undocumented but still supported for
469 backwards compatibility. It is equivalent to /readonly except that in
470 ACLs where /readonly is required you may specify /leaky/noupdate or
471 /strict/noupdate which are treated the same as /readonly.
473 A useful new feature is the /count= option. This is a generalization
474 of the per_byte option, so that you can measure the throughput of other
475 aggregate values. For example, the per_byte option is now equivalent
476 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
478 The per_rcpt option has been generalized using the /count= mechanism
479 (though it's more complicated than the per_byte equivalence). When it is
480 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
481 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
482 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
483 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
484 Note that using per_rcpt with a non-readonly update mode in more than
485 one ACL will cause the recipients to be double-counted. (The per_mail
486 and per_byte options don't have this problem.)
488 The handling of very low rates has changed slightly. If the computed rate
489 is less than the event's count (usually one) then this event is the first
490 after a long gap. In this case the rate is set to the same as this event's
491 count, so that the first message of a spam run is counted properly.
493 The major new feature is a mechanism for counting the rate of unique
494 events. The new per_addr option counts the number of different
495 recipients that someone has sent messages to in the last time period. It
496 behaves like per_rcpt if all the recipient addresses are different, but
497 duplicate recipient addresses do not increase the measured rate. Like
498 the /count= option this is a general mechanism, so the per_addr option
499 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
500 example, measure the rate that a client uses different sender addresses
501 with the options per_mail/unique=$sender_address. There are further
502 details in the main documentation.
504 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
506 TF/06 Removed a few PCRE remnants.
508 TF/07 Automatically extract Exim's version number from tags in the git
509 repository when doing development or release builds.
511 PP/02 Raise smtp_cmd_buffer_size to 16kB.
512 Bugzilla 879. Patch from Paul Fisher.
514 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
515 Heavily based on revision 40f9a89a from Simon Arlott's tree.
518 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
520 PP/05 Variable $av_failed, true if the AV scanner deferred.
521 Bugzilla 1078. Patch from John Horne.
523 PP/06 Stop make process more reliably on build failure.
524 Bugzilla 1087. Patch from Heiko Schlittermann.
526 PP/07 Make maildir_use_size_file an _expandable_ boolean.
527 Bugzilla 1089. Patch from Heiko Schlittermann.
529 PP/08 Handle ${run} returning more data than OS pipe buffer size.
530 Bugzilla 1131. Patch from Holger Weiß.
532 PP/09 Handle IPv6 addresses with SPF.
533 Bugzilla 860. Patch from Wolfgang Breyha.
535 PP/10 GnuTLS: support TLS 1.2 & 1.1.
537 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
540 PP/11 match_* no longer expand right-hand-side by default.
541 New compile-time build option, EXPAND_LISTMATCH_RHS.
542 New expansion conditions, "inlist", "inlisti".
544 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
546 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
548 PP/14 fix log_write() format string regression from TF/03.
549 Bugzilla 1152. Patch from Dmitry Isaikin.
555 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
557 PP/02 Harmonised TLS library version reporting; only show if debugging.
558 Layout now matches that introduced for other libraries in 4.74 PP/03.
560 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
562 PP/04 New "dns_use_edns0" global option.
564 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
567 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
568 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
570 TK/01 Updated PolarSSL code to 0.14.2.
571 Bugzilla 1097. Patch from Andreas Metzler.
573 PP/07 Catch divide-by-zero in ${eval:...}.
576 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
579 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
580 format-string attack -- SECURITY: remote arbitrary code execution.
582 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
583 time unintentionally subject to list matching rules, letting the header
584 cause arbitrary Exim lookups (of items which can occur in lists, *not*
585 arbitrary string expansion). This allowed for information disclosure.
587 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
588 INT_MIN/-1 -- value coerced to INT_MAX.
594 NM/01 Workround for PCRE version dependency in version reporting
597 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
598 This fixes portability to compilers other than gcc, notably
599 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
601 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
602 makefiles for portability to HP-UX and POSIX correctness.
604 PP/01 Permit LOOKUP_foo enabling on the make command-line.
605 Also via indented variable definition in the Makefile.
606 (Debugging by Oliver Heesakkers).
608 PP/02 Restore caching of spamd results with expanded spamd_address.
609 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
611 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
612 Improves build reliability. Fix from: Frank Elsner
614 NM/02 Fix wide character breakage in the rfc2047 coding
615 Fixes bug 1064. Patch from Andrey N. Oktyabrski
617 NM/03 Allow underscore in dnslist lookups
618 Fixes bug 1026. Patch from Graeme Fowler
620 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
621 Code patches from Adam Ciarcinski of NetBSD.
623 NM/04 Fixed exiqgrep to cope with mailq missing size issue
626 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
627 is logged, to avoid truncation. Patch from John Horne.
629 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
630 Patch from Jakob Hirsch.
632 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
633 SQL string expansion failure details.
634 Patch from Andrey Oktyabrski.
636 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
637 Patch from Simon Arlott.
639 PP/09 New lookups functionality failed to compile on old gcc which rejects
640 extern declarations in function scope.
641 Patch from Oliver Fleischmann
643 PP/10 Use sig_atomic_t for flags set from signal handlers.
644 Check getgroups() return and improve debugging.
645 Fixed developed for diagnosis in bug 927 (which turned out to be
648 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
649 Patch from Mark Zealey.
651 PP/12 Bugzilla 1056: Improved spamd server selection.
652 Patch from Mark Zealey.
654 PP/13 Bugzilla 1086: Deal with maildir quota file races.
655 Based on patch from Heiko Schlittermann.
657 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
658 Patch from Uwe Doering, sign-off by Michael Haardt.
660 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
661 variable declaration deep within a block. Bug and patch from
664 PP/15 lookups-Makefile IRIX compatibilty coercion.
666 PP/16 Make DISABLE_DKIM build knob functional.
668 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
669 Patch by Simon Arlott
671 TF/03 Fix valgrind.h portability to C89 compilers that do not support
672 variable argument macros. Our copy now differs from upstream.
678 TF/01 Failure to get a lock on a hints database can have serious
679 consequences so log it to the panic log.
681 TF/02 Log LMTP confirmation messages in the same way as SMTP,
682 controlled using the smtp_confirmation log selector.
684 TF/03 Include the error message when we fail to unlink a spool file.
686 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
687 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
688 for maintaining out-of-tree patches for some time.
690 PP/01 Bugzilla 139: Documentation and portability issues.
691 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
692 Handle per-OS dynamic-module compilation flags.
694 PP/02 Let /dev/null have normal permissions.
695 The 4.73 fixes were a little too stringent and complained about the
696 permissions on /dev/null. Exempt it from some checks.
697 Reported by Andreas M. Kirchwitz.
699 PP/03 Report version information for many libraries, including
700 Exim version information for dynamically loaded libraries. Created
701 version.h, now support a version extension string for distributors
702 who patch heavily. Dynamic module ABI change.
704 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
705 privilege escalation vulnerability whereby the Exim run-time user
706 can cause root to append content of the attacker's choosing to
709 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
712 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
713 If dropping privileges for untrusted macros, we disabled normal logging
714 on the basis that it would fail; for the Exim run-time user, this is not
715 the case, and it resulted in successful deliveries going unlogged.
716 Fixed. Reported by Andreas Metzler.
722 PP/01 Date: & Message-Id: revert to normally being appended to a message,
723 only prepend for the Resent-* case. Fixes regression introduced in
724 Exim 4.70 by NM/22 for Bugzilla 607.
726 PP/02 Include check_rfc2047_length in configure.default because we're seeing
727 increasing numbers of administrators be bitten by this.
729 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
731 PP/03 Bugzilla 994: added openssl_options main configuration option.
733 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
735 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
737 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
739 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
740 without a peer certificate, leading to a segfault because of an
741 assumption that peers always have certificates. Be a little more
742 paranoid. Problem reported by Martin Tscholak.
744 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
745 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
746 NB: ClamAV planning to remove STREAM in "middle of 2010".
747 CL also introduces -bmalware, various -d+acl logging additions and
748 more caution in buffer sizes.
750 PP/09 Implemented reverse_ip expansion operator.
752 PP/10 Bugzilla 937: provide a "debug" ACL control.
754 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
756 PP/12 Bugzilla 973: Implement --version.
758 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
760 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
762 PP/15 Bugzilla 816: support multiple condition rules on Routers.
764 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
765 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
766 ignore trailing whitespace.
768 JJ/02 prevent non-panic DKIM error from being sent to paniclog
770 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
773 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
774 Notification from Dr Andrew Aitchison.
776 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
777 ExtendedDetectionInfo response format.
778 Notification from John Horne.
780 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
783 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
784 XSL and documented dependency on system catalogs, with examples of how
787 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
790 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
791 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
792 configuration file which is writeable by the Exim user or group.
794 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
795 of configuration files to cover files specified with the -C option if
796 they are going to be used with root privileges, not just the default
799 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
800 option (effectively making it always true).
802 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
803 files to be used while preserving root privileges.
805 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
806 that rogue child processes cannot use them.
808 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
809 run-time user, instead of root.
811 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
812 Exim run-time user without dropping privileges.
814 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
815 result string, instead of calling string_vformat() twice with the same
818 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
819 for other users. Others should always drop root privileges if they use
820 -C on the command line, even for a whitelisted configure file.
822 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
824 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
830 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
831 $data_path, and $header_path variables; fixed documentation bugs and
834 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
835 exipick to access non-standard spools, including the "frozen" queue
838 NM/01 Bugzilla 965: Support mysql stored procedures.
839 Patch from Alain Williams
841 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
843 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
844 Patch from Andreas Metzler
846 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
847 Patch from Kirill Miazine
849 NM/05 Bugzilla 671: Added umask to procmail example.
851 JJ/03 installed exipick 20100323.0, fixing doc bug
853 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
854 directory. Notification and patch from Dan Rosenberg.
856 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
858 TK/02 Improve log output when DKIM signing operation fails.
860 MH/01 Treat the transport option dkim_domain as a colon separated
861 list, not as a single string, and sign the message with each element,
862 omitting multiple occurences of the same signer.
864 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
865 Bugzilla 985, 986. Patch by Simon Arlott
867 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
868 Patch by Simon Arlott
870 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
871 MBX locking. Notification from Dan Rosenberg.
877 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
879 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
881 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
883 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
885 NM/04 Bugzilla 915: Flag broken perl installation during build.
891 TK/01 Added patch by Johannes Berg that expands the main option
892 "spamd_address" if it starts with a dollar sign.
894 TK/02 Write list of recipients to X-Envelope-Sender header when building
895 the mbox-format spool file for content scanning (suggested by Jakob
898 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
899 (http://www.dcc-servers.net/) support via dccifd. Activated by
900 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
902 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
903 by Mark Daniel Reidel <mr@df.eu>.
905 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
906 When building exim an external PCRE library is now needed -
907 PCRE is a system library on the majority of modern systems.
908 See entry on PCRE_LIBS in EDITME file.
910 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
911 conversation. Added nologin parameter to request.
912 Patch contributed by Kirill Miazine.
914 TF/01 Do not log submission mode rewrites if they do not change the address.
916 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
918 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
919 log files in place. Contributed by Roberto Lima.
921 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
923 TF/03 Bugzilla 615: When checking the local_parts router precondition
924 after a local_part_suffix or local_part_prefix option, Exim now
925 does not use the address's named list lookup cache, since this
926 contains cached lookups for the whole local part.
928 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
929 Robert Millan. Documentation is in experimental-spec.txt.
931 TF/04 Bugzilla 668: Fix parallel build (make -j).
933 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
935 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
936 Patch provided by Jan Srzednicki.
938 TF/05 Leading white space used to be stripped from $spam_report which
939 wrecked the formatting. Now it is preserved.
941 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
942 that they are available at delivery time.
944 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
946 TF/08 TLS error reporting now respects the incoming_interface and
947 incoming_port log selectors.
949 TF/09 Produce a more useful error message if an SMTP transport's hosts
950 setting expands to an empty string.
952 NM/06 Bugzilla 744: EXPN did not work under TLS.
953 Patch provided by Phil Pennock.
955 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
956 Patch provided by Richard Godbee.
958 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
959 acl_smtp_notquit, added index entry.
961 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
962 Patch provided by Eugene Bujak.
964 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
965 accept(). Patch provided by Maxim Dounin.
967 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
968 Patch provided by Phil Pennock.
970 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
972 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
973 Patch provided by Brad "anomie" Jorsch.
975 NM/14 Bugzilla 622: Added timeout setting to transport filter.
976 Patch provided by Dean Brooks.
978 TK/05 Add native DKIM support (does not depend on external libraries).
980 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
981 Patch provided by Graeme Fowler.
983 NM/16 Bugzilla 851: Documentation example syntax fix.
985 NM/17 Changed NOTICE file to remove references to embedded PCRE.
987 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
990 NM/19 Bugzilla 745: TLS version reporting.
991 Patch provided by Phil Pennock.
993 NM/20 Bugzilla 167: bool: condition support.
994 Patch provided by Phil Pennock.
996 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
997 clients. Patch provided by Phil Pennock.
999 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1000 Patch provided by Brad "anomie" Jorsch.
1002 NM/23 Bugzilla 687: Fix misparses in eximstats.
1003 Patch provided by Heiko Schlittermann.
1005 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1006 Patch provided by Heiko Schlittermann.
1008 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1009 plus update to original patch.
1011 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1013 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1014 Patch provided by David Brownlee.
1016 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1018 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1020 NM/30 Bugzilla 888: TLS documentation bugfixes.
1022 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1024 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1025 Unlike the original bugzilla I have changed all shell scripts in src tree.
1027 NM/33 Bugzilla 898: Transport filter timeout fix.
1028 Patch by Todd Rinaldo.
1030 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1031 Patch by Serge Demonchaux.
1033 NM/35 Bugzilla 39: Base64 decode bug fixes.
1034 Patch by Jakob Hirsch.
1036 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1038 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1040 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1042 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1048 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1049 ALT-N's libdkim that I have put here:
1050 http://duncanthrax.net/exim-experimental/
1052 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1053 were called 'true' and it seems that C99 defines that as a reserved
1054 keyword to be used with 'bool' variable types. That means you could
1055 not include C99-style headers which use bools without triggering
1056 build errors in sieve.c.
1058 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1059 as mailq or other aliases. Changed the --help handling significantly
1060 to do whats expected. exim_usage() emits usage/help information.
1062 SC/01 Added the -bylocaldomain option to eximstats.
1064 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1066 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1068 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1074 PH/01 Another patch from the Sieve maintainer.
1076 PH/02 When an IPv6 address is converted to a string for single-key lookup
1077 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1078 dots are used instead of colons so that keys in lsearch files need not
1079 contain colons. This was done some time before quoting was made available
1080 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1081 (notated using the quote facility) so as to distinguish them from IPv4
1082 keys. This meant that lookups for IP addresses in host lists did not work
1083 for iplsearch lookups.
1085 This has been fixed by arranging for IPv6 addresses to be expressed with
1086 colons if the lookup type is iplsearch. This is not incompatible, because
1087 previously such lookups could never work.
1089 The situation is now rather anomolous, since one *can* have colons in
1090 ordinary lsearch keys. However, making the change in all cases is
1091 incompatible and would probably break a number of configurations.
1093 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1096 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1097 conversion specification without a maximum field width, thereby enabling
1098 a rogue spamd server to cause a buffer overflow. While nobody in their
1099 right mind would setup Exim to query an untrusted spamd server, an
1100 attacker that gains access to a server running spamd could potentially
1101 exploit this vulnerability to run arbitrary code as the Exim user.
1103 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1104 $primary_hostname instead of what libspf2 thinks the hosts name is.
1106 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1107 a directory entry by the name of the lookup key. Previously, if a
1108 symlink pointed to a non-existing file or a file in a directory that
1109 Exim lacked permissions to read, a lookup for a key matching that
1110 symlink would fail. Now it is enough that a matching directory entry
1111 exists, symlink or not. (Bugzilla 503.)
1113 PH/03 The body_linecount and body_zerocount variables are now exported in the
1116 PH/04 Added the $dnslist_matched variable.
1118 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1119 This means they are set thereafter only if the connection becomes
1122 PH/06 Added the client_condition to authenticators so that some can be skipped
1123 by clients under certain conditions.
1125 PH/07 The error message for a badly-placed control=no_multiline_responses left
1126 "_responses" off the end of the name.
1128 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1130 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1131 (without spaces) instead of just copying the configuration text.
1133 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1135 PH/11 Added $max_received_linelength.
1137 PH/12 Added +ignore_defer and +include_defer to host lists.
1139 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1140 way in which PCRE > 7.0 is built.
1142 PH/14 Implemented queue_only_load_latch.
1144 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1145 MAIL command. The effect was to mangle the value on 64-bit systems.
1147 PH/16 Another patch from the Sieve maintainer.
1149 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1151 PH/18 If a system quota error occurred while trying to create the file for
1152 a maildir delivery, the message "Mailbox is full" was not appended to the
1153 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1154 only to a quota excession during the actual writing of the file.
1156 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1157 characters?) which causes problems in log lines. The DN values are now
1158 passed through string_printing() before being added to log lines.
1160 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1161 and InterBase are left for another time.)
1163 PH/21 Added message_body_newlines option.
1165 PH/22 Guard against possible overflow in moan_check_errorcopy().
1167 PH/23 POSIX allows open() to be a macro; guard against that.
1169 PH/24 If the recipient of an error message contained an @ in the local part
1170 (suitably quoted, of course), incorrect values were put in $domain and
1171 $local_part during the evaluation of errors_copy.
1177 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1178 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1181 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1182 issue a MAIL command.
1184 PH/02 In an ACL statement such as
1186 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1188 if a client was not listed at all, or was listed with a value other than
1189 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1190 the condition was not true (as it should be), so access was not denied.
1191 The bug was that the ! inversion was incorrectly passed on to the second
1192 item. This has been fixed.
1194 PH/03 Added additional dnslists conditions == and =& which are different from
1195 = and & when the dns lookup returns more than one IP address.
1197 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1198 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1200 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1201 FSYNC, which compiles an option called disable_fsync that allows for
1202 bypassing fsync(). The documentation is heavily laced with warnings.
1204 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1206 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1207 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1208 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1209 including adding "make clean"; (3) Added -fPIC when compiling the test
1210 dynamically loaded module, to get rid of a warning.
1212 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1213 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1214 = 0s. The bug is otherwise harmless.
1216 PH/07 There was a bug in the dovecot authenticator such that the value of
1217 $auth1 could be overwritten, and so not correctly preserved, after a
1218 successful authentication. This usually meant that the value preserved by
1219 the server_setid option was incorrect.
1221 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1223 PH/09 Installed PCRE release 7.0.
1225 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1226 run for batched SMTP input. It is now run at the start of every message
1227 in the batch. While fixing this I discovered that the process information
1228 (output by running exiwhat) was not always getting set for -bs and -bS
1229 input. This is fixed, and it now also says "batched" for BSMTP.
1231 PH/11 Added control=no_pipelining.
1233 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1234 patch, slightly modified), and move the expansion of helo_data till after
1235 the connection is made in the smtp transport (so it can use these
1238 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1240 PH/14 Added log_selector = +pid.
1242 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1244 PH/16 Add ${if forany and ${if forall.
1246 PH/17 Added dsn_from option to vary the From: line in DSNs.
1248 PH/18 Flush SMTP output before performing a callout, unless control =
1249 no_callout_flush is set.
1251 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1252 was true (the default) a successful delivery failed to delete the retry
1253 item, thus causing premature timeout of the address. The bug is now
1256 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1258 PH/21 Long custom messages for fakedefer and fakereject are now split up
1259 into multiline reponses in the same way that messages for "deny" and
1260 other ACL rejections are.
1262 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1263 with slight modification.
1265 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1266 draft, changing the syntax and factoring some duplicate code".
1268 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1269 for deliveries of the second and subsequent messages over the same SMTP
1272 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1273 ${reduce, with only minor "tidies".
1275 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1277 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1278 expansion side effects.
1280 PH/27 When a message times out after an over-quota error from an Exim-imposed
1281 quota, the bounce message says "mailbox is full". This message was not
1282 being given when it was a system quota that was exceeded. It now should
1285 MH/03 Made $recipients available in local_scan(). local_scan() already has
1286 better access to the recipient list through recipients_list[], but
1287 $recipients can be useful in postmaster-provided expansion strings.
1289 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1290 in the case of a MAIL command with additional options following the
1291 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1292 were accidentally chopped off.
1294 PH/29 SMTP synchronization checks are implemented when a command is read -
1295 there is a check that no more input is waiting when there shouldn't be
1296 any. However, for some commands, a delay in an ACL can mean that it is
1297 some time before the response is written. In this time, more input might
1298 arrive, invalidly. So now there are extra checks after an ACL has run for
1299 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1300 pipelining has not been advertised.
1302 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1304 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1305 returned in the SMTP response when smtp_return_error_details was false.
1306 This has been fixed.
1308 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1309 instead of the C I/O that was originally supplied, because problems were
1310 reported on Solaris.
1312 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1313 Exim which did not show up earlier: it was assuming that a call to
1314 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1315 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1316 was a macro that became an assignment, so it seemed to work. This has
1317 changed to a proper function call with a void return, hence the compile
1318 error. Exim's code has been fixed.
1320 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1323 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1325 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1327 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1328 because it turns out to be too big or there is a timeout) there is no
1329 "Completed" line in the log. When some messages of this type were
1330 selected by exigrep, they were listed as "not completed". Others were
1331 picked up by some special patterns. I have improved the selection
1332 criteria to be more general.
1334 PH/38 The host_find_failed option in the manualroute router can now be set
1335 to "ignore", to completely ignore a host whose IP address cannot be
1336 found. If all hosts are ignored, the behaviour is controlled by the new
1337 host_all_ignored option.
1339 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1340 homing or because of multiple MX records with /mx) generated more than
1341 one IP address, and the following item turned out to be the local host,
1342 all the secondary addresses of the first item were incorrectly removed
1343 from the list, along with the local host and any following hosts (which
1344 is what is supposed to happen).
1346 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1347 whoever called Exim into the -H file. In the case of the daemon it was
1348 behaving confusingly. When first started, it used values for whoever
1349 started the daemon, but after a SIGHUP it used the Exim user (because it
1350 calls itself on a restart). I have changed the code so that it now always
1353 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1354 message are rejected with the same error (e.g. no authentication or bad
1355 sender address), and a DATA command is nevertheless sent (as can happen
1356 with PIPELINING or a stupid MUA), the error message that was given to the
1357 RCPT commands is included in the rejection of the DATA command. This is
1358 intended to be helpful for MUAs that show only the final error to their
1361 PH/42 Another patch from the Sieve maintainer.
1363 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1364 Eximstats - Fixed some broken HTML links and added missing column headers
1366 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1367 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1369 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1371 PH/43 Yet another patch from the Sieve maintainer.
1373 PH/44 I found a way to check for a TCP/IP connection going away before sending
1374 the response to the final '.' that terminates a message, but only in the
1375 case where the client has not sent further data following the '.'
1376 (unfortunately, this is allowed). However, in many cases there won't be
1377 any further data because there won't be any more messages to send. A call
1378 to select() can be used: if it shows that the input is "ready", there is
1379 either input waiting, or the socket has been closed. An attempt to read
1380 the next input character can distinguish the two cases. Previously, Exim
1381 would have sent an OK response which the client would never have see.
1382 This could lead to message repetition. This fix should cure that, at
1383 least in a lot of common cases.
1385 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1386 advertised in response to EHLO.
1392 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1393 fixed by 4.65/MH/01 (is this a record?) are fixed:
1395 (i) An empty string was always treated as zero by the numeric comparison
1396 operators. This behaviour has been restored.
1398 (ii) It is documented that the numeric comparison operators always treat
1399 their arguments as decimal numbers. This was broken in that numbers
1400 starting with 0 were being interpreted as octal.
1402 While fixing these problems I realized that there was another issue that
1403 hadn't been noticed. Values of message_size_limit (both the global option
1404 and the transport option) were treated as octal if they started with 0.
1405 The documentation was vague. These values are now always treated as
1406 decimal, and I will make that clear in the documentation.
1412 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1413 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1416 MH/01 Don't check that the operands of numeric comparison operators are
1417 integers when their expansion is in "skipping" mode (fixes bug
1418 introduced by 4.64-PH/07).
1420 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1421 child addresses, Exim now panics and dies. Previously, because the count
1422 is held in a short int, deliveries were likely to be lost. As such a
1423 large number of recipients for a single message is ridiculous
1424 (performance will be very, very poor), I have chosen to impose a limit
1425 rather than extend the field.
1431 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1432 leftover -K file (the existence of which was triggered by #402).
1433 While we were at it, introduced process PID as part of the -K
1434 filename. This should rule out race conditions when creating
1437 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1438 processing considerably. Previous code took too long for large mails,
1439 triggering a timeout which in turn triggers #401.
1441 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1442 in the DK code in transports.c. sendfile() is not really portable,
1443 hence the _LINUX specificness.
1445 TF/01 In the add_headers option to the mail command in an Exim filter,
1446 there was a bug that Exim would claim a syntax error in any
1447 header after the first one which had an odd number of characters
1450 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1451 callout verification, Exim cached a "reject" for the entire domain. This
1452 is correct for most verifications, but it is not correct for a recipient
1453 verification with use_sender or use_postmaster set, because in that case
1454 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1455 case of MAIL FROM:<> rejection from other early rejections (e.g.
1456 rejection of HELO). When verifying a recipient using a non-null MAIL
1457 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1458 Whatever the result of the callout, the value of the domain cache is
1459 left unchanged (for any other kind of callout, getting as far as trying
1460 RCPT means that the domain itself is ok).
1462 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
1465 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
1466 manifest itself as EPIPE rather than ECONNECT. When tidying away a
1467 session, the daemon ignores ECONNECT errors and logs others; it now
1468 ignores EPIPE as well.
1470 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
1471 (quoted-printable decoding).
1473 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
1474 later the small subsequent patch to fix an introduced bug.
1476 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
1478 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
1480 PH/08 An error is now given if message_size_limit is specified negative.
1482 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
1483 to be given (somewhat) arbitrary names.
1485 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
1488 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
1489 miscellaneous code fixes
1491 PH/10 Added the log_reject_target ACL modifier to specify where to log
1494 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
1495 hostname. This is wrong, because it relates to the incoming message (and
1496 probably the interface on which it is arriving) and not to the outgoing
1497 callout (which could be using a different interface). This has been
1498 changed to use the value of the helo_data option from the smtp transport
1499 instead - this is what is used when a message is actually being sent. If
1500 there is no remote transport (possible with a router that sets up host
1501 addresses), $smtp_active_hostname is used.
1503 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
1504 tweaks were necessary in order to get it to work (see also 21 below):
1505 (a) The code assumed that strncpy() returns a negative number on buffer
1506 overflow, which isn't the case. Replaced with Exim's string_format()
1508 (b) There were several signed/unsigned issues. I just did the minimum
1509 hacking in of casts. There is scope for a larger refactoring.
1510 (c) The code used strcasecmp() which is not a standard C function.
1511 Replaced with Exim's strcmpic() function.
1512 (d) The code set only $1; it now sets $auth1 as well.
1513 (e) A simple test gave the error "authentication client didn't specify
1514 service in request". It would seem that Dovecot has changed its
1515 interface. Fortunately there's a specification; I followed it and
1516 changed what the client sends and it appears to be working now.
1518 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
1521 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
1522 address A is aliased to B and C, where B exists and C does not. Without
1523 -v the output is "A verified" because verification stops after a
1524 successful redirection if more than one address is generated. However,
1525 with -v the child addresses are also verified. Exim was outputting "A
1526 failed to verify" and then showing the successful verification for C,
1527 with its parentage. It now outputs "B failed to verify", showing B's
1528 parentage before showing the successful verification of C.
1530 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
1531 look up a TXT record in a specific list after matching in a combined
1534 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
1535 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
1536 they consult the DNS. I had assumed they would set it the way they
1537 wanted; and indeed my experiments on Linux seem to show that in some
1538 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
1539 To be on the safe side, however, I have now made the interface to
1540 host_find_byname() similar to host_find_bydns(), with an argument
1541 containing the DNS resolver options. The host_find_byname() function now
1542 sets these options at its start, just as host_find_bydns() does. The smtp
1543 transport options dns_qualify_single and dns_search_parents are passed to
1544 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
1545 of host_find_byname() use the default settings of RES_DEFNAMES
1546 (qualify_single) but not RES_DNSRCH (search_parents).
1548 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
1549 spool_read_header() do less string testing, by means of a preliminary
1550 switch on the second character of optional "-foo" lines. (This is
1551 overdue, caused by the large number of possibilities that now exist.
1552 Originally there were few.) While I was there, I also converted the
1553 str(n)cmp tests so they don't re-test the leading "-" and the first
1554 character, in the hope this might squeeze out yet more improvement.
1556 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
1557 flag allowing group syntax was set by the header_syntax check but not
1558 turned off, possible causing trouble later; (2) The flag was not being
1559 set at all for the header_verify test, causing "group"-style headers to
1560 be rejected. I have now set it in this case, and also caused header_
1561 verify to ignore an empty address taken from a group. While doing this, I
1562 came across some other cases where the code for allowing group syntax
1563 while scanning a header line wasn't quite right (mostly, not resetting
1564 the flag correctly in the right place). These bugs could have caused
1565 trouble for malformed header lines. I hope it is now all correct.
1567 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
1568 with the "reply" argument non-NULL. The code, however (which originally
1569 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
1570 but it didn't always do it. This confused somebody who was copying the
1571 code for some other use. I have removed all the tests.
1573 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
1574 feature that was used to support insecure browsers during the U.S. crypto
1575 embargo. It requires special client support, and Exim is probably the
1576 only MTA that supported it -- and would never use it because real RSA is
1577 always available. This code has been removed, because it had the bad
1578 effect of slowing Exim down by computing (never used) parameters for the
1579 RSA_EXPORT functionality.
1581 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
1582 authenticator to fail if there's a tab character in the incoming data
1583 (there should never be unless someone is messing about, as it's supposed
1584 to be base64-encoded). Also added, on Timo's advice, the "secured" option
1585 if the connection is using TLS or if the remote IP is the same as the
1586 local IP, and the "valid-client-cert option" if a client certificate has
1589 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
1590 authenticators. This can be used for authorization after authentication
1591 succeeds. (In the case of plaintext, it servers for both authentication
1594 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
1595 if any retry times were supplied.
1597 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
1598 connection, where there is no client IP address to check. In this
1599 situation, the verify now always succeeds.
1601 PH/25 Applied John Jetmore's -Mset patch.
1603 PH/26 Added -bem to be like -Mset, but loading a message from a file.
1605 PH/27 In a string expansion for a processed (not raw) header when multiple
1606 headers of the same name were present, leading whitespace was being
1607 removed from all of them, but trailing whitespace was being removed only
1608 from the last one. Now trailing whitespace is removed from each header
1609 before concatenation. Completely empty headers in a concatenation (as
1610 before) are ignored.
1612 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
1613 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
1615 PH/29 [Removed. This was a change that I later backed out, and forgot to
1616 correct the ChangeLog entry (that I had efficiently created) before
1617 committing the later change.]
1619 PH/30 Exim was sometimes attempting to deliver messages that had suffered
1620 address errors (4xx response to RCPT) over the same connection as other
1621 messages routed to the same hosts. Such deliveries are always "forced",
1622 so retry times are not inspected. This resulted in far too many retries
1623 for the affected addresses. The effect occurred only when there were more
1624 hosts than the hosts_max_try setting in the smtp transport when it had
1625 the 4xx errors. Those hosts that it had tried were not added to the list
1626 of hosts for which the message was waiting, so if all were tried, there
1627 was no problem. Two fixes have been applied:
1629 (i) If there are any address or message errors in an SMTP delivery, none
1630 of the hosts (tried or untried) are now added to the list of hosts
1631 for which the message is waiting, so the message should not be a
1632 candidate for sending over the same connection that was used for a
1633 successful delivery of some other message. This seems entirely
1634 reasonable: after all the message is NOT "waiting for some host".
1635 This is so "obvious" that I'm not sure why it wasn't done
1636 previously. Hope I haven't missed anything, but it can't do any
1637 harm, as the worst effect is to miss an optimization.
1639 (ii) If, despite (i), such a delivery is accidentally attempted, the
1640 routing retry time is respected, so at least it doesn't keep
1641 hammering the server.
1643 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
1644 in ${readsocket because some servers need this prod.
1646 PH/32 Added some extra debug output when updating a wait-xxx database.
1648 PH/33 The hint "could be header name not terminated by colon", which has been
1649 given for certain expansion errors for a long time, was not being given
1650 for the ${if def:h_colon_omitted{... case.
1652 PH/34 The spec says: "With one important exception, whenever a domain list is
1653 being scanned, $domain contains the subject domain." There was at least
1654 one case where this was not true.
1656 PH/35 The error "getsockname() failed: connection reset by peer" was being
1657 written to the panic log as well as the main log, but it isn't really
1658 panic-worthy as it just means the connection died rather early on. I have
1659 removed the panic log writing for the ECONNRESET error when getsockname()
1662 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
1663 runs only) independently of the message's sender address. This meant
1664 that, if the 4xx error was in fact related to the sender, a different
1665 message to the same recipient with a different sender could confuse
1666 things. In particualar, this can happen when sending to a greylisting
1667 server, but other circumstances could also provoke similar problems.
1668 I have changed the default so that the retry time for these errors is now
1669 based a combination of the sender and recipient addresses. This change
1670 can be overridden by setting address_retry_include_sender=false in the
1673 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
1674 remote server are returned as part of bounce messages. This was not
1675 happening for LMTP over a pipe (the lmtp transport), but now it is the
1676 same for both kinds of LMTP.
1678 PH/38 Despite being documented as not happening, Exim was rewriting addresses
1679 in header lines that were in fact CNAMEs. This is no longer the case.
1681 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
1682 and queue runs started by the daemon processed all messages. This has
1683 been fixed so that -R and -S can now usefully be given with -q<time>.
1685 PH/40 Import PCRE release 6.7 (fixes some bugs).
1687 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
1689 PH/42 Give an error if -q is specified more than once.
1691 PH/43 Renamed the variables $interface_address and $interface_port as
1692 $received_ip_address and $received_port, to make it clear that these
1693 values apply to message reception, and not to the outgoing interface when
1694 a message is delivered. (The old names remain recognized, of course.)
1696 PH/44 There was no timeout on the connect() call when using a Unix domain
1697 socket in the ${readsocket expansion. There now is.
1699 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
1700 be meaningful with "accept".
1702 SC/01 Eximstats V1.43
1703 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
1705 SC/02 Eximstats V1.44
1706 Use a glob alias rather than an array ref in the generated
1707 parser. This improves both readability and performance.
1709 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
1710 Collect SpamAssassin and rejection statistics.
1711 Don't display local sender or destination tables unless
1712 there is data to show.
1713 Added average volumes into the top table text output.
1715 SC/04 Eximstats V1.46
1716 Collect data on the number of addresses (recipients)
1717 as well as the number of messages.
1719 SC/05 Eximstats V1.47
1720 Added 'Message too big' to the list of mail rejection
1721 reasons (thanks to Marco Gaiarin).
1723 SC/06 Eximstats V1.48
1724 Mainlog lines which have GMT offsets and are too short to
1725 have a flag are now skipped.
1727 SC/07 Eximstats V1.49 (Alain Williams)
1728 Added the -emptyok flag.
1730 SC/08 Eximstats V1.50
1731 Fixes for obtaining the IP address from reject messages.
1733 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
1734 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
1735 whitesspace changes from 4.64-PH/27
1737 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
1740 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
1741 are found, allow negative numbers in numeric criteria)
1743 JJ/06 exipick.20061117.2, added new $message_body_missing variable
1745 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
1746 to match changes made in 4.64-PH/43
1748 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
1750 PH/47 Put in an explicit test for a DNS lookup of an address record where the
1751 "domain" is actually an IP address, and force a failure. This locks out
1752 those revolvers/nameservers that support "A-for-A" lookups, in
1753 contravention of the specifications.
1755 PH/48 When a host name was looked up from an IP address, and the subsequent
1756 forward lookup of the name timed out, the host name was left in
1757 $sender_host_name, contrary to the specification.
1759 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
1760 restricted to single-key lookups, Exim was not diagnosing an error if
1761 * or *@ was used with a query-style lookup.
1763 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
1765 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
1766 long ago, but noone interested enough thought of it. Let's just say that
1767 the "1.1" means that there are some new functions that weren't there at
1768 some point in the past.
1770 PH/51 Error processing for expansion failure of helo_data from an smtp
1771 transport during callout processing was broken.
1773 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
1774 tested/used via the -bh/-bhc/-bs options.
1776 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
1777 bug, fixed in subsequent PCRE releases).
1779 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
1780 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
1782 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
1788 SC/01 Use a glob alias rather than an array ref in eximstats generated
1789 parser. This improves both readability and performance.
1791 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
1792 Don't display local sender or destination tables in eximstats unless
1793 there is data to show.
1794 Added average volumes into the eximstats top table text output.
1796 SC/03 Collect data on the number of addresses (recipients) as well
1797 as the number of messages in eximstats.
1799 TF/01 Correct an error in the documentation for the redirect router. Exim
1800 does (usually) call initgroups() when daemonizing.
1802 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
1803 with consistent privilege compared to when running as a daemon.
1805 TF/03 Note in the spec that $authenticated_id is not set for local
1806 submissions from trusted users.
1808 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
1809 Thanks to Dean Brooks <dean@iglou.com> for the patch.
1811 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
1812 by adding some example configuration directives to the default
1813 configuration file. A little bit of work is required to uncomment the
1814 directives and define how usernames and passwords are checked, but
1815 there is now a framework to start from.
1817 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
1818 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
1819 without this. I don't know how relevant this is to other LDAP libraries.
1821 PH/02 Add the verb name to the "unknown ACL verb" error.
1823 PH/03 Magnus Holmgren's patch for filter_prepend_home.
1825 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
1827 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
1828 directory not expanded when it should be if an expanded home directory
1829 was set for the address (which is overridden by the transport).
1831 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
1834 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
1835 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
1836 because it is too late at that time, and has no effect.
1838 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
1839 security issue with \' (bugzilla #107). I could not use the
1840 PQescapeStringConn() function, because it needs a PGconn value as one of
1843 PH/08 When testing addresses using -bt, indicate those final addresses that
1844 are duplicates that would not cause an additional delivery. At least one
1845 person was confused, thinking that -bt output corresponded to deliveries.
1846 (Suppressing duplicates isn't a good idea as you lose the information
1847 about possibly different redirections that led to the duplicates.)
1849 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
1850 systems where poll() doesn't work, in particular OS X.
1852 PH/10 Added more information to debugging output for retry time not reached.
1854 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
1855 operations in malware.c.
1857 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
1860 PH/13 If write_rejectlog was set false when logging was sent to syslog with
1861 syslog_duplication set false, log lines that would normally be written
1862 both the the main log and to the reject log were not written to syslog at
1865 PH/14 In the default configuration, change the use of "message" in ACL warn
1866 statements to "add_header".
1868 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
1869 not followed by a command (e.g. "seen endif").
1871 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
1872 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
1875 PH/17 Added extra conditions to the default value of delay_warning_condition
1879 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
1880 { match{$h_precedence:}{(?i)bulk|list|junk} } \
1881 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
1884 The Auto-Submitted: and various List- headers are standardised, whereas I
1885 don't think Precedence: ever was.
1887 PH/18 Refactored debugging code in route_finduser() to show more information,
1888 in particular, the error code if getpwnam() issues one.
1890 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
1891 This is apparently needed in addition to the PH/07 change above to avoid
1892 any possible encoding problems.
1894 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
1895 but not after initializing Perl.
1897 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
1898 output them only if debugging. By default they are written stderr,
1899 apparently, which is not desirable.
1901 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
1904 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
1907 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1909 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1910 authenticated or an ident call has been made. Suppress the default
1911 values for $authenticated_id and $authenticated_sender (but permit -oMai
1912 and -oMas) when testing with -bh.
1914 PH/24 Re-jigged the order of the tests in the default configuration so that the
1915 tests for valid domains and recipients precede the DNS black list and CSA
1916 tests, on the grounds that those ones are more expensive.
1918 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1919 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1920 This bug exists in every version of Exim that I still have, right back to
1923 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1924 However, an attempt to turn on case-sensitivity in a regex key by
1925 including (?-i) didn't work because the subject string was already
1926 lowercased, and the effects were non-intuitive. It turns out that a
1927 one-line patch can be used to allow (?-i) to work as expected.
1933 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1934 other effects) broke the use of negated acl sub-conditions.
1936 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1939 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1940 "Deny" causes Exim to reject the incoming connection with a 554 error.
1941 Unfortunately, if there is a major crisis, such as a disk failure,
1942 tcp-wrappers gives "deny", whereas what one would like would be some
1943 kind of temporary error. A kludge has been added to help with this.
1944 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1945 554 error is used if errno is still zero or contains ENOENT (which occurs
1946 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1949 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1951 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1952 errors. Otherwise a message that provokes a temporary error (when other
1953 messages do not) can cause a whole host to time out.
1955 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1956 addresses were routed directly to files or pipes from a redirect router.
1957 File deliveries just didn't batch; pipe deliveries might have suffered
1960 PH/06 A failure to get a lock for a hints database would erroneously always say
1961 "Failed to get write lock", even when it was really a read lock.
1963 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1964 of 0600. This has been changed to use the value of the lockfile_mode
1965 option (which defaults to 0600).
1967 PH/08 Applied small patch from the Sieve maintainer.
1969 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1970 folder from quota calculations, a direct delivery into this folder messed
1971 up the contents of the maildirsize file. This was because the regex was
1972 used only to exclude .Trash (or whatever) when the size of the mailbox
1973 was calculated. There was no check that a delivery was happening into an
1974 excluded directory. This bug has been fixed by ignoring all quota
1975 processing for deliveries into excluded directories.
1977 PH/10 Added the maildirfolder_create_regex option to appendfile.
1983 PH/01 The code for finding all the local interface addresses on a FreeBSD
1984 system running IPv6 was broken. This may well have applied to all BSD
1985 systems, as well as to others that have similar system calls. The broken
1986 code found IPv4 interfaces correctly, but gave incorrect values for the
1987 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1988 that it would not match correctly against @[] and not recognize the IPv6
1991 PH/02 The ipliteral router was not recognizing addresses of the form user@
1992 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1994 PH/03 Added disable_ipv6.
1996 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1997 decoded form, because it is most often used to construct To: headers
1998 lines in autoreplies, and the decoded form may well be syntactically
1999 invalid. However, $reply_address has leading white space removed, and all
2000 newlines turned into spaces so that the autoreply transport does not
2003 PH/05 If group was specified without a user on a router, and no group or user
2004 was specified on a transport, the group from the router was ignored.
2006 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2007 for visible compile-time settings that can be used to change these
2008 numbers, for those that want even more. Backwards compatibility with old
2009 spool files has been maintained. However, going back to a previous Exim
2010 release will lost any variables that are in spool files.
2012 PH/07 Two small changes when running in the test harness: increase delay when
2013 passing a TCP/IP connection to a new process, in case the original
2014 process has to generate a bounce, and remove special handling of
2015 127.0.0.2 (sic), which is no longer necessary.
2017 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2018 be the same on different OS.
2020 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2023 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2024 whether --show-vars was specified or not
2026 JJ/02 exipick: Added support for new ACL variable spool format introduced
2029 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2030 syntactically invalid From: or Reply-to: line, and a filter used this to
2031 generate an autoreply, and therefore failed to obtain an address for the
2032 autoreply, Exim could try to deliver to a non-existent relative file
2033 name, causing unrelated and misleading errors. What now happens is that
2034 it logs this as a hard delivery error, but does not attempt to create a
2037 PH/11 The exinext utility has a -C option for testing purposes, but although
2038 the given file was scanned by exinext itself; it wasn't being passed on
2039 when Exim was called.
2041 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2042 an end-of-file indication when reading a command response.
2044 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2045 compiled. In many other places in Exim, IPv6 addresses are always
2046 recognized, so I have changed this. It also means that IPv4 domain
2047 literals of the form [IPV4:n.n.n.n] are now always recognized.
2049 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2050 used if the router is not running as root, for example, when verifying at
2051 ACL time, or when using -bh. The debugging output from this situation was
2052 non-existent - all you got was a failure to exec. I have made two
2055 (a) Failures to set uid/gid, the current directory, or a process leader
2056 in a subprocess such as that created by queryprogram now generate
2057 suitable debugging ouput when -d is set.
2059 (b) The queryprogram router detects when it is not running as root,
2060 outputs suitable debugging information if -d is set, and then runs
2061 the subprocess without attempting to change uid/gid.
2063 PH/15 Minor change to Makefile for building test_host (undocumented testing
2066 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2067 additional section of a DNS packet that returns MX or SRV records.
2068 Instead, it always explicitly searches for A/AAAA records. This avoids
2069 major problems that occur when a DNS server includes only records of one
2070 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2071 fixed another bug: if SRV records were looked up and the corresponding
2072 address records were *not* found in the additional section, the port
2073 values from the SRV records were lost.
2075 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2076 using the correct key (the original address) when searching the retry
2077 rules in order to find which one to use for generating the retry hint.
2079 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2080 adding the default one. Similarly, if it contains a Reply-To: header, the
2081 errors_reply_to option, if set, is not used.
2083 PH/19 When calculating a retry time, Exim used to measure the "time since
2084 failure" by looking at the "first failed" field in the retry record. Now
2085 it does not use this if it is later than than the arrival time of the
2086 message. Instead it uses the arrival time. This makes for better
2087 behaviour in cases where some deliveries succeed, thus re-setting the
2088 "first failed" field. An example is a quota failure for a huge message
2089 when small messages continue to be delivered. Without this change, the
2090 "time since failure" will always be short, possible causing more frequent
2091 delivery attempts for the huge message than are intended.
2092 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2094 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2095 $1, $2, $3) because the numerical variables can be reset during some
2096 expansion items (e.g. "match"), thereby losing the authentication data.
2098 PH/21 Make -bV show the size of off_t variables so that the test suite can
2099 decide whether to run tests for quotas > 2G.
2101 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2102 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2103 filecount value is greater than 2G or if a quota value is greater than 2G
2104 on a system where the size of off_t is not greater than 4, a panic error
2107 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2108 never match. The debug and -bh output now contains an explicit error
2109 message indicating a malformed IPv4 address or mask.
2111 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2112 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2113 PH/23 above applies.
2115 PH/25 Do not write to syslog when running in the test harness. The only
2116 occasion when this arises is a failure to open the main or panic logs
2117 (for which there is an explicit test).
2119 PH/26 Added the /no_tell option to "control=freeze".
2121 PH/27 If a host name lookup failed very early in a connection, for example, if
2122 the IP address matched host_lookup and the reverse lookup yielded a name
2123 that did not have a forward lookup, an error message of the form "no IP
2124 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2125 could be logged. Now it outputs the IP address instead of "NULL".
2127 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2128 allows the sender and the authenticated sender to be set when
2129 submitting a message from within Exim. Since child_open_exim() is
2130 documented for local_scan(), the new function should be too.
2132 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2133 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2134 results in an empty string is now treated as unset.
2136 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2138 PH/31 Added sender_verify_fail logging option.
2140 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2141 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2142 tidied the source and removed it altogether.
2144 PH/33 When a queue run was abandoned because the load average was too high, a
2145 log line was always written; now it is written only if the queue_run log
2146 selector is set. In addition, the log line for abandonment now contains
2147 information about the queue run such as the pid. This is always present
2148 in "start" and "stop" lines but was omitted from the "abandon" line.
2150 PH/34 Omit spaces between a header name and the colon in the error message that
2151 is given when verify = headers_syntax fails (if there are lots of them,
2152 the message gets confusing).
2154 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2155 names, as there are now some PTR records that contain slashes. This check
2156 is only to protect against broken name servers that fall over on strange
2157 characters, so the fact that it applies to all lookups doesn't matter.
2159 PH/36 Now that the new test suite is complete, we can remove some of the
2160 special code in Exim that was needed for the old test suite. For example,
2161 sorting DNS records because real resolvers return them in an arbitrary
2162 order. The new test suite's fake resolver always returns records in the
2165 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2166 bounces) except when queue_only is set, to avoid logging races between
2167 the different processes.
2169 PH/38 Panic-die if .include specifies a non-absolute path.
2171 PH/39 A tweak to the "H" retry rule from its user.
2173 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2174 a label. They prevented compilation on older perls.
2176 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2177 a warning to be raised on newish perls.
2179 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2180 on queue. Changes to match documented behaviour of showing count of
2181 messages matching specified criteria.
2183 PH/40 Changed the default ident timeout from 30s to 5s.
2185 PH/41 Added support for the use of login_cap features, on those BSD systems
2186 that have them, for controlling the resources used by pipe deliveries.
2188 PH/42 The content-scanning code uses fopen() to create files in which to put
2189 message data. Previously it was not paying any attention to the mode of
2190 the files. Exim runs with umask(0) because the rest of the code creates
2191 files with open(), and sets the required mode explicitly. Thus, these
2192 files were ending up world-writeable. This was not a big issue, because,
2193 being within the spool directory, they were not world-accessible. I have
2194 created a function called modefopen, which takes an additional mode
2195 argument. It sets umask(777), creates the file, chmods it to the required
2196 mode, then resets the umask. All the relevant calls to fopen() in the
2197 content scanning code have been changed to use this function.
2199 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2200 to 24 hours. This avoids potential overflow problems when processing G
2201 and H retry rules. I suspect nobody ever tinkers with this value.
2203 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2205 PH/45 When the plaintext authenticator is running as a client, the server's
2206 challenges are checked to ensure they are valid base64 strings. By
2207 default, the authentication attempt is cancelled if an invalid string is
2208 received. Setting client_ignore_invalid_base64 true ignores these errors.
2209 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2210 they are received. Thus, the responses can be made to depend on the
2211 challenges. If an invalid string is ignored, an empty string is placed in
2214 PH/46 Messages that are created by the autoreply transport now contains a
2215 References: header, in accordance with RFCs 2822 and 3834.
2217 PH/47 Added authenticated_sender_force to the smtp transport.
2219 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2221 PH/49 Installed latest patch from the Sieve maintainer.
2223 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2224 was also set, the appendfile transport was unnecessarily scanning a
2225 directory of message files (e.g. for maildir delivery) to find the count
2226 of files (along with the size), even though it did not need this
2227 information. It now does the scan only if it needs to find either the
2228 size of the count of files.
2230 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2232 PH/52 Two bugs concerned with error handling when the smtp transport is
2235 (i) Exim was not creating retry information for temporary errors given
2236 for individual recipients after the DATA command when the smtp transport
2237 was used in LMTP mode. This meant that they could be retried too
2238 frequently, and not timed out correctly.
2240 (ii) Exim was setting the flag that allows error details to be returned
2241 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2242 recipients that were returned after the DATA command.
2244 PH/53 This is related to PH/52, but is more general: for any failing address,
2245 when detailed error information was permitted to be returned to the
2246 sender, but the error was temporary, then after the final timeout, only
2247 "retry timeout exceeded" was returned. Now it returns the full error as
2248 well as "retry timeout exceeded".
2250 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2251 do this, and (what is worse) MTAs that accept it.
2253 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2254 will now be deprecated.
2256 PH/56 New os.c-cygwin from the Cygwin maintainer.
2258 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2259 formats (previously only available in exim formats via -bpr, -bpru,
2260 and -bpra. Now also available in native and exiqgrep formats)
2262 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2263 with very large, slow to parse queues
2265 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2267 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2269 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2270 responses to authentication challenges, though it was showing the
2271 challenges; (ii) I've removed the CR characters from the debug output for
2274 PH/58 Allow for the insertion of a newline as well as a space when a string
2275 is turned into more than one encoded-word during RFC 2047 encoding. The
2276 Sieve code now uses this.
2278 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2279 data_4xx, lost_connection, tls_required.
2281 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2282 message was being sent as an SMTP response.
2284 PH/61 Add -l and -k options to exicyclog.
2286 PH/62 When verifying, if an address was redirected to one new address, so that
2287 verification continued, and the new address failed or deferred after
2288 having set something in $address_data, the value of $address_data was not
2289 passed back to the ACL. This was different to the case when no
2290 redirection occurred. The value is now passed back in both cases.
2292 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2293 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2294 use login_cap.h, so on its own it isn't the distinguishing feature. The
2295 new name refers directly to the setclassresources() function.
2297 PH/65 Added configuration files for NetBSD3.
2299 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2301 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2302 is preferred over IPv4.
2304 PH/68 The bounce_return_message and bounce_return_body options were not being
2305 honoured for bounces generated during the reception of non-SMTP messages.
2306 In particular, this applied to messages rejected by the ACL. This bug has
2307 been fixed. However, if bounce_return_message is true and bounce_return_
2308 body is false, the headers that are returned for a non-SMTP message
2309 include only those that have been read before the error was detected.
2310 (In the case of an ACL rejection, they have all been read.)
2312 PH/69 The HTML version of the specification is now built in a directory called
2313 spec_html instead of spec.html, because the latter looks like a path with
2314 a MIME-type, and this confuses some software.
2316 PH/70 Catch two compiler warnings in sieve.c.
2318 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2319 function verify_get_ident() calls ip_connect() to connect a socket, but
2320 if the "connect()" function timed out, ip_connect() used to close the
2321 socket. However, verify_get_ident() also closes the socket later, and in
2322 between Exim writes to the log, which may get opened at this point. When
2323 the socket was closed in ip_connect(), the log could get the same file
2324 descriptor number as the socket. This naturally causes chaos. The fix is
2325 not to close the socket in ip_connect(); the socket should be closed by
2326 the function that creates it. There was only one place in the code where
2327 this was missing, in the iplookup router, which I don't think anybody now
2328 uses, but I've fixed it anyway.
2330 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2331 well as to direct DNS lookups. Otherwise the handling of names in host
2332 lists is inconsistent and therefore confusing.
2338 PH/01 Two changes to the default runtime configuration:
2340 (1) Move the checks for relay_from_hosts and authenticated clients from
2341 after to before the (commented out) DNS black list checks.
2343 (2) Add control=submission to the relay_from_hosts and authenticated
2344 clients checks, on the grounds that messages accepted by these
2345 statements are most likely to be submissions.
2347 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2349 (1) Generate an error if the third argument for the ${prvs expansion is
2352 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2355 (3) Reset the variables that are obtained from the first argument of
2356 ${prvscheck and used in the second argument before leaving the code,
2357 because their memory is reclaimed, so using them afterwards may do
2360 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2361 one (it's much easier than Tom thought :-).
2363 (5) Because of (4), we can now allow for the use of $prvscheck_result
2364 inside the third argument.
2366 PH/03 For some reason, the default setting of PATH when running a command from
2367 a pipe transport was just "/usr/bin". I have changed it to
2370 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2371 anything to be listed in the output from -bV.
2373 PH/05 When a filter generated an autoreply, the entire To: header line was
2374 quoted in the delivery log line, like this:
2376 => >A.N.Other <ano@some.domain> <original@ddress> ...
2378 This has been changed so that it extracts the operative address. There
2379 may be more than one such address. If so, they are comma-separated, like
2382 => >ano@some.domain,ona@other.domain <original@ddress> ...
2384 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2385 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2386 not being looked up in the rDNS to get a host name, Exim was showing the
2387 IP address twice in Received: lines, even though the IP addresses were
2388 identical. For example:
2390 Received: from [1.2.3.4] (helo=[1.2.3.4])
2392 However, if the real host name was known, it was omitting the HELO data
2393 if it matched the actual IP address. This has been tidied up so that it
2394 doesn't show the same IP address twice.
2396 PH/07 When both +timestamp and +memory debugging was on, the value given by
2397 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2398 called by the debug printing, thereby overwriting the timestamp buffer.
2399 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2401 PH/08 When the original message was included in an autoreply transport, it
2402 always said "this is a copy of the message, including all the headers",
2403 even if body_only or headers_only was set. It now gives an appropriate
2406 PH/09 Applied a patch from the Sieve maintainer which:
2408 o fixes some comments
2409 o adds the (disabled) notify extension core
2410 o adds some debug output for the result of if/elsif tests
2411 o points to the current vacation draft in the documentation
2412 and documents the missing references header update
2416 o fixes a bug in processing the envelope test (when testing
2417 multiple envelope elements, the last element determinted the
2420 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2421 Electronic Mail") by including:
2423 Auto-submitted: auto-generated
2425 in the messages that it generates (bounce messages and others, such as
2426 warnings). In the case of bounce messages for non-SMTP mesages, there was
2427 also a typo: it was using "Auto_submitted" (underscore instead of
2428 hyphen). Since every message generated by Exim is necessarily in response
2429 to another message, thes have all been changed to:
2431 Auto-Submitted: auto-replied
2433 in accordance with these statements in the RFC:
2435 The auto-replied keyword:
2437 - SHOULD be used on messages sent in direct response to another
2438 message by an automatic process,
2440 - MUST NOT be used on manually-generated messages,
2442 - MAY be used on Delivery Status Notifications (DSNs) and Message
2443 Disposition Notifications (MDNs),
2445 - MUST NOT be used on messages generated by automatic or periodic
2446 processes, except for messages which are automatic responses to
2449 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2450 to the default Received: header definition.
2452 PH/12 Added log selector acl_warn_skipped (default on).
2454 PH/13 After a successful wildlsearch lookup, discard the values of numeric
2455 variables because (a) they are in the wrong storage pool and (b) even if
2456 they were copied, it wouldn't work properly because of the caching.
2458 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2459 checking when decoding. Apparently there are clients that generate
2460 overlong encoded strings. Why am I not surprised?
2462 PH/15 If the first argument of "${if match_address" was not empty, but did not
2463 contain an "@" character, Exim crashed. Now it writes a panic log message
2464 and treats the condition as false.
2466 PH/16 In autoreply, treat an empty string for "once" the same as unset.
2468 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
2469 extension "envelope-auth". The code is finished and in agreement with
2470 other implementations, but there is no documentation so far and in fact,
2471 nobody wrote the draft yet. This extension is currently #undef'ed, thus
2472 not changing the active code.
2474 Print executed "if" and "elsif" statements when debugging is used. This
2475 helps a great deal to understand what a filter does.
2477 Document more things not specified clearly in RFC3028. I had all this
2478 sorted out, when out of a sudden new issues came to my mind. Oops."
2480 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
2483 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
2484 canonical form (as documented). However, after a host name lookup from
2485 the IP address, check_host() was doing a simple string comparison with
2486 addresses acquired from the DNS when checking that the found name did
2487 have the original IP as one of its addresses. Since any found IPv6
2488 addresses are likely to be in abbreviated form, the comparison could
2489 fail. Luckily, there already exists a function for doing the comparison
2490 by converting both addresses to binary, so now that is used instead of
2491 the text comparison.
2493 PH/20 There was another similar case to PH/19, when a complete host name was
2494 given in a host list; looking up its IP address could give an abbreviated
2495 form, whereas the current host's name might or might not be abbreviated.
2496 The same fix has been applied.
2502 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
2503 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
2506 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
2507 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
2509 PH/03 Typo: missing ".o" in src/pcre/Makefile.
2511 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
2512 header line, restrict the check to what is listed in RFCs 2369 and 2929.
2513 Also, for "Auto-Submitted", treat anything other than "no" as
2514 non-personal, in accordance with RFC 3834. (Previously it treated
2515 anything starting "auto-" as non-personal.)
2517 TF/01 The control=submission/name=... option had a problem with syntax
2518 errors if the name included a slash character. The /name= option
2519 now slurps the rest of the string, so it can include any characters
2520 but it must come last in the list of options (after /sender_retain
2523 PH/05 Some modifications to the interface to the fake nameserver for the new
2531 TK/01 Added the "success_on_redirect" address verification option. See
2532 NewStuff for rationale and an example.
2534 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
2536 PH/02 Patch to exigrep to allow it to work on syslog lines.
2538 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
2539 fread() to skip over the body file's header line, because in Cygwin the
2540 header line is locked and is inaccessible.
2542 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
2543 co-exist for some time) to make it clear that it is the Exim ID that is
2544 referenced, not the Message-ID: header line.
2546 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
2547 string_format() function, because snprintf() does not exist on all
2550 PH/06 The use of forbid_filter_existstest now also locks out the use of the
2551 ${stat: expansion item.
2553 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
2554 protocol synchronization error", to keep the pedants happy.
2556 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
2557 well as for IRIX systems, when gcc is being used. See the host.c source
2560 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
2562 PH/10 Named domain lists were not working if used in a queue_smtp_domains
2565 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
2566 transport and to the smtp transport in LMTP mode.
2568 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
2570 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
2571 run a filter in a subprocess. This could lead to confusion in subsequent
2572 lookups in the parent process. There should also be a search_tidyup() at
2573 the end of the subprocess.
2575 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
2576 only if the host matched helo_try_verify_hosts, which caused the
2577 verification to occur when the EHLO/HELO command was issued. The ACL just
2578 tested the remembered result. Now, if a previous verification attempt has
2579 not happened, "verify = helo" does it there and then.
2581 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
2583 TK/03 Fix log output including CR from clamd.
2585 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
2586 exist provoked a memory error which could cause a segfault.
2588 PH/15 Installed PCRE 6.2
2590 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
2592 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
2593 of the problem. Specifically, suggested +O2 rather than +O1 for the
2596 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
2598 PH/20 If a delivery was routed to a non-standard port by means of an SRV
2599 record, the port was not correctly logged when the outgoing_port log
2600 selector was set (it logged the transort's default port).
2602 PH/21 Added support for host-specific ports to manualroute, queryprogram,
2603 fallback_hosts, and "hosts" in the smtp transport.
2605 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
2606 host errors such as "Connection refused".
2608 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
2609 authentication with radiusclient 0.4.9:
2611 - Error returned from rc_read_config was caught wrongly
2612 - Username/password not passed on to radius server due to wrong length.
2614 The presumption is that some radiusclient API changes for 4.51/PH/17
2615 were not taken care of correctly. The code is still untested by me (my
2616 Linux distribution still has 0.3.2 of radiusclient), but it was
2617 contributed by a Radius user.
2619 PH/24 When doing a callout, the value of $domain wasn't set correctly when
2620 expanding the "port" option of the smtp transport.
2622 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
2623 while reading a MIME header. Thanks to Tom Hughes for a patch.
2625 PH/24 Include config.h inside local_scan.h so that configuration settings are
2628 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
2629 that in an ACL for RCPT (for example), you can examine exactly what was
2632 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
2633 commands, but it was not correctly comparing the address with the actual
2634 client host address. Thus, it would show the EHLO address in Received:
2635 header lines when this was not necessary.
2637 PH/27 Added the % operator to ${eval:}.
2639 PH/28 Exim tries to create and chdir to its spool directory when it starts;
2640 it should be ignoring failures (because with -C, for example, it has lost
2641 privilege). It wasn't ignoring creation failures other than "already
2644 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
2647 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
2648 because an input line was too long, either on its own, or by virtue of
2649 too many continuations, the temporary file was not being removed, and the
2650 return code was incorrect.
2652 PH/31 Missing "BOOL" in function definition in filtertest.c.
2654 PH/32 Applied Sieve patches from the maintainer.
2656 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
2658 PH/33 Added "verify = not_blind".
2660 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
2661 Local/Makefile (with some defaults set). These are used in built scripts
2662 such as exicyclog, but they have never been used in the exim_install
2663 script (though there are many overriding facilities there). I have
2664 arranged that the exim_install script now takes note of these two
2667 PH/35 Installed configuration files for Dragonfly.
2669 PH/36 When a locally submitted message by a trusted user did not contain a
2670 From: header, and the sender address was obtained from -f or from an SMTP
2671 MAIL command, and the trusted user did not use -F to supply a sender
2672 name, $originator_name was incorrectly used when constructing a From:
2673 header. Furthermore, $originator_name was used for submission mode
2674 messages from external hosts without From: headers in a similar way,
2675 which is clearly wrong.
2677 PH/37 Added control=suppress_local_fixups.
2679 PH/38 When log_selector = +received_sender was set, and the addition of the
2680 sender made the log line's construction buffer exactly full, or one byte
2681 less than full, an overflow happened when the terminating "\n" was
2684 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
2685 when the result of a list match is failure because a DNS lookup failed.
2687 PH/40 RM_COMMAND is now used in the building process.
2689 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
2690 the "build-* directories that it finds.
2692 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
2693 address in a domain literal was a prefix of an interface address.
2695 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
2696 when verifying a sender address, unless rewrite_headers is false.
2698 PH/44 Wrote a long comment about why errors_to addresses are verified as
2699 recipients, not senders.
2701 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
2702 the ratelimit ACL was added.
2704 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
2706 PH/46 Added extra information about PostgreSQL errors to the error string.
2708 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
2709 avoiding the need to install special zones in a real server. This is
2710 backwards compatible; if it can't find the fake resolver, it drops back.
2711 Thus, both old and new test suites can be run.
2713 TF/02 Added util/ratelimit.pl
2715 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
2716 clock is set back in time.
2718 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
2719 Candler <B.Candler@pobox.com>.
2721 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
2722 OK for addresses that are the result of redirections.
2724 PH/48 A number of further additions for the benefit of the new test suite,
2725 including a fake gethostbyname() that interfaces to the fake DNS resolver
2728 TF/06 The fix for widen_domains has also been applied to qualify_single and
2729 search_parents which are the other dnslookup options that can cause
2732 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
2735 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
2737 TF/07 Exim produced the error message "an SRV record indicated no SMTP
2738 service" if it encountered an MX record with an empty target hostname.
2739 The message is now "an MX or SRV record indicated no SMTP service".
2741 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
2742 if the DNS of the sending site is misconfigured. This is quite a
2743 common situation. This change restores the behaviour of treating a
2744 helo verification defer as a failure.
2746 PH/51 If self=fail was set on a router, the bounce message did not include the
2747 actual error message.
2753 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
2755 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
2756 command itself ended in error, the underlying message about the transport
2757 filter timeout was being overwritten with the pipe command error. Now the
2758 underlying error message should be appended to the second error message.
2760 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
2762 PH/02 Reduce the amount of output that "make" produces by default. Full output
2763 can still be requested.
2765 PH/03 The warning log line about a condition test deferring for a "warn" verb
2766 was being output only once per connection, rather than after each
2767 occurrence (because it was using the same function as for successful
2768 "warn" verbs). This seems wrong, so I have changed it.
2770 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
2771 it should not have, which might have caused a crash in the right
2772 circumstances, but probably never did.
2774 PH/04 Installed a modified version of Tony Finch's patch to make submission
2775 mode fix the return path as well as the Sender: header line, and to
2776 add a /name= option so that you can make the user's friendly name appear
2779 TF/03 Added the control = fakedefer ACL modifier.
2781 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
2782 Mark Lowes for thorough testing.
2784 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
2786 TK/03 Merged latest SRS patch from Miles Wilton.
2788 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
2789 with the definition in sysexits.h (which is #included earlier).
2790 Fortunately, Exim does not actually use EX_OK. The code used to try to
2791 preserve the sysexits.h value, by assumimg that macro definitions were
2792 scanned for macro replacements. I have been disabused of this notion,
2793 so now the code just undefines EX_OK before #including unistd.h.
2795 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
2796 in the smtp transport. When a block could not be written in a single
2797 write() function, the timeout was being re-applied to each part-write.
2798 This seems wrong - if the receiver was accepting one byte at a time it
2799 would take for ever. The timeout is now adjusted when this happens. It
2800 doesn't have to be particularly precise.
2802 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
2803 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
2805 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
2806 without a domain if the check to <postmaster@domain> fails.
2808 SC/01 Eximstats: added -xls and the ability to specify output files
2809 (patch written by Frank Heydlauf).
2811 SC/02 Eximstats: use FileHandles for outputing results.
2813 SC/03 Eximstats: allow any combination of xls, txt, and html output.
2815 SC/04 Eximstats: fixed display of large numbers with -nvr option
2817 SC/05 Eximstats: fixed merging of reports with empty tables.
2819 SC/06 Eximstats: added the -include_original_destination flag
2821 SC/07 Eximstats: removed tabs and trailing whitespace.
2823 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
2825 TK/06 MBOX spool code: Add real "From " MBOX separator line
2826 so the .eml file is really in mbox format (even though
2827 most programs do not really care). Patch from Alex Miller.
2829 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
2830 The latter is generated from $received_to and is only set if the
2831 message has one envelope recipient. SA can use these headers,
2832 obviously out-of-the-box. Patch from Alex Miller.
2834 PH/08 The ${def test on a variable was returning false if the variable's
2835 value was "0", contrary to what the specification has always said!
2836 The result should be true unless the variable is empty.
2838 PH/09 The syntax error of a character other than { following "${if
2839 def:variable_name" (after optional whitespace) was not being diagnosed.
2840 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
2841 accidental colon was present, for example, could give incorrect results.
2843 PH/10 Tidied the code in a number of places where the st_size field of a stat()
2844 result is used (not including appendfile, where other changes are about
2847 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
2848 This involved changing a lot of size variables from int to off_t. It
2849 should work with maildirs and everything.
2851 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
2852 spamd dying while we are connected to it.
2854 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
2857 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
2858 function for BDB 4.3.
2860 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
2862 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
2863 the sources for PCRE have been re-arranged and more files are now
2866 PH/15 The code I had for printing potentially long long variables in PH/11
2867 above was not the best (it lost precision). The length of off_t variables
2868 is now inspected at build time, and an appropriate printing format (%ld
2869 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
2870 to be "long long int" or "long int". This is needed for the internal
2871 formatting function string_vformat().
2873 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
2874 the configuration file to be ":syslog", then the script "guesses" where
2875 the logs files are, rather than using the compiled in default. In our
2876 case the guess is not the same as the compiled default, so the script
2877 suddenly stopped working when I started to use syslog. The patch checks
2878 to see if log_file_path is "". If so, it attempts to read it from exim
2879 with no configuration file to get the compiled in version, before it
2880 falls back to the previous guessing code."
2882 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
2883 implementing BATV in an Exim configuration. See NewStuff for the gory
2886 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
2887 Makefile that are specific to HP-UX.
2889 PH/18 If the "use_postmaster" option was set for a recipient callout together
2890 with the "random" option, the postmaster address was used as the MAIL
2891 FROM address for the random test, but not for the subsequent recipient
2892 test. It is now used for both.
2894 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
2895 patch removes a few documentation additions to RFC 3028, because the
2896 latest draft now contains them. It adds the new en;ascii-case comparator
2897 and a new error check for 8bit text in MIME parts. Comparator and
2898 require names are now matched exactly. I enabled the subaddress
2899 extension, but it is not well tested yet (read: it works for me)."
2901 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
2902 rework some of the code of TK/09 above to avoid the hardwired use of
2903 "%lld" and "long long". Replaced the call to snprintf() with a call to
2906 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
2907 records point to non-existent hosts", "retry timeout exceeded", and
2908 "retry time not reached for any host after a long failure period".
2910 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2911 experimental DomainKeys support:
2913 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2914 (2) On an error such as an illegally used "control", the wrong name for
2915 the control was given.
2917 These problems did NOT occur unless DomainKeys support was compiled.
2919 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2921 PH/24 Added ${if match_ip condition.
2923 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2924 databases so that it will be absolutely obvious if a crash occurs in the
2925 DB library. This is a regular occurrence (often caused by mis-matched
2928 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2929 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2930 picked up on a user's system that detects such things. There doesn't seem
2931 to be a gcc warning option for this - only an attribute that has to be
2932 put on the function's prototype. It seems that in Fedora Core 4 they have
2933 set this on a number of new functions. No doubt there will be more in due
2936 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2937 specify a transport. However, if an address that was verified by such a
2938 router was the subject of a callout, Exim crashed because it tried to
2939 read the rcpt_include_affixes from the non-existent transport. Now it
2940 just assumes that the setting of that option is false. This bug was
2941 introduced by 4.51/PH/31.
2943 PH/28 Changed -d+all to exclude +memory, because that information is very
2944 rarely of interest, but it makes the output a lot bigger. People tend to
2945 do -d+all out of habit.
2947 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2948 code in os-type was giving problems when libc.so lives in lib64, like on
2951 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2952 aren't the modern standard, and it seems that some systems' include files
2953 don't always have them. Exim was already checking for some of the newer
2954 ones like T_AAAA, and defining it itself. I've added checks for all the
2955 record types that Exim uses.
2957 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2958 not automatically generating a new one, as it is supposed to. This
2959 prevented TLS from working. If the file did exist, but contained invalid
2960 data, a new version was generated, as expected. It was only the case of a
2961 non-existent file that was broken.
2963 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2964 with a change in libdomainkeys > 0.64.
2966 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2967 from DNS. If the selector record carries the flag, it now has
2968 precedence over the domain-wide flag.
2970 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2972 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2973 the use of a port name that isn't defined in /etc/services) occurred, the
2974 message was deferred as in a normal delivery, and thus remained on the
2975 spool, instead of being failed because of the mua_wrapper setting. This
2976 is now fixed, and I tidied up some of the mua_wrapper messages at the
2979 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2980 the messages in a hash of arrays rather than using individual hashes.
2981 This is a bit cleaner and results in dramatic memory savings, albeit
2982 at a slight CPU cost.
2984 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2985 as requested by Marc Sherman.
2987 SC/10 Eximstats: added histograms for user specified patterns as requested
2990 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2992 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2993 fopen() in the content-scanning modules that did not already have it.
2999 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3000 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3002 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3004 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3006 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3007 to test Sieve filters that use "vacation".
3009 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3010 that changes the way the GnuTLS parameters are stored in the cache file.
3011 The new format can be generated externally. For backward compatibility,
3012 if the data in the cache doesn't make sense, Exim assumes it has read an
3013 old-format file, and it generates new data and writes a new file. This
3014 means that you can't go back to an older release without removing the
3017 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3018 work if there are any delivery delays because "one_time" forces the
3019 parent to be marked "delivered", so its unseen clone is never tried
3020 again. For this reason, Exim now forbids the simultaneous setting of
3023 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3024 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3025 entry if you want to know the details. The fix, however, neglected to
3026 consider the case when local delivery batching is involved. The test for
3027 "previously delivered" was not happening when checking to see if an
3028 address could be batched with a previous (undelivered) one; under
3029 certain circumstances this could lead to multiple deliveries to the same
3032 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3033 in its include files, and this causes problems building Exim.
3035 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3036 header_syntax) but Exim was just ignoring anything given after a slash.
3037 In particular, this caused confusion with an attempt to use "verify =
3038 reverse_host_lookup/defer_ok". An error is now given when options are
3039 supplied for verify items that do not have them. (Maybe reverse_host_
3040 lookup should have a defer_ok option, but that's a different point.)
3042 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3043 defined by RFC 821) to 2048, because there were problems with some AUTH
3044 commands, and RFC 1869 says the size should be increased for extended
3045 SMTP commands that take arguments.
3047 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3050 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3051 "unknown" error; now it says that the functionality isn't in the binary.
3053 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3054 an address' error message when a string expansion fails (syntax or
3055 whatever). Otherwise the password may appear in the log. Following change
3056 PH/42 below, there is no longer a chance of it appearing in a bounce
3059 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3061 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3062 of its addresses was ever tried. (Bugzilla bug #2.)
3064 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3065 the result incorrectly in the debug output. (It correctly added a newline
3066 to what was transported.)
3068 TF/01 Added $received_time.
3070 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3071 commented out examples of how to interface to a virus scanner and to
3072 SpamAssassin. Also added commented examples of av_scanner and
3073 spamd_address settings.
3075 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3076 and controls are allowed in which ACLs. There were a couple of minor
3077 errors. Some of the entries in the conditions table (which is a table of
3078 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3079 negation of where the condition IS allowed.
3081 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3083 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3084 header file does not have a version number, so I've had to invent a new
3085 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3086 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3087 radiusclient), but it was contributed by a Radius user.
3089 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3090 files or directories, for OpenSSL.
3092 PH/19 When an Exim process that is running as root has to create an Exim log
3093 file, it does so in a subprocess that runs as exim:exim so as to get the
3094 ownership right at creation (otherwise, other Exim processes might see
3095 the file with the wrong ownership). There was no test for failure of this
3096 fork() call, which would lead to the process getting stuck as it waited
3097 for a non-existent subprocess. Forks do occasionally fail when resources
3098 run out. I reviewed all the other calls to fork(); they all seem to check
3101 PH/20 When checking for unexpected SMTP input at connect time (before writing
3102 the banner), Exim was not dealing correctly with a non-positive return
3103 from the read() function. If the client had disconnected by this time,
3104 the result was a log entry for a synchronization error with an empty
3105 string after "input=" when read() returned zero. If read() returned -1
3106 (an event I could not check), uninitialized data bytes were printed.
3107 There were reports of junk text (parts of files, etc) appearing after
3110 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3112 PH/22 Added support for macro redefinition, and (re)definition in between
3113 driver and ACL definitions.
3115 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3116 forgetting to use the resulting value; it was using the unexpanded value.
3118 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3119 hadn't been configured. The fix is from Juergen Kreileder, who
3120 understands it better than I do:
3122 "Here's what I see happening with three configured cyrus_sasl
3123 authenticators configured (plain, login, cram-md5):
3125 On startup auth_cyrus_sasl_init() gets called for each of these.
3126 This means three calls to sasl_listmech() without a specified mech_list.
3127 => SASL tests which mechs of all available mechs actually work
3128 => three warnings about OTP not working
3129 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3131 With the patch, sasl_listmech() also gets called three times. But now
3132 SASL's mech_list option is set to the server_mech specified in the the
3133 authenticator. Or in other words, the answer from sasl_listmech()
3134 gets limited to just the mech you're testing for (which is different
3136 => the return list contains just 'plain' or 'login', 'cram-md5' or
3137 nothing depending on the value of ob->server_mech.
3139 I've just tested the patch: Authentication still works fine,
3140 unavailable mechs specified in the exim configuration are still
3141 caught, and the auth.log warnings about OTP are gone."
3143 PH/25 When debugging is enabled, the contents of the command line are added
3144 to the debugging output, even when log_selector=+arguments is not
3147 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3148 answer is "GNU", and only if the return is "GNU/something" is the answer
3151 PH/27 $acl_verify_message is now set immediately after the failure of a
3152 verification in an ACL, and so is available in subsequent modifiers. In
3153 particular, the message can be preserved by coding like this:
3155 warn !verify = sender
3156 set acl_m0 = $acl_verify_message
3158 Previously, $acl_verify_message was set only while expanding "message"
3159 and "log_message" when a very denied access.
3161 PH/28 Modified OS/os.c-Linux with
3163 -#ifndef OS_LOAD_AVERAGE
3164 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3166 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3167 nomenclature these days.)
3169 PH/29 Installed patch from the Sieve maintainer that adds the options
3170 sieve_useraddress and sieve_subaddress to the redirect router.
3172 PH/30 In these circumstances:
3173 . Two addresses routed to the same list of hosts;
3174 . First host does not offer TLS;
3175 . First host accepts first address;
3176 . First host gives temporary error to second address;
3177 . Second host offers TLS and a TLS session is established;
3178 . Second host accepts second address.
3179 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3180 and peerdn, if requested) that were in fact used only for the second
3183 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3184 attention to any local part prefix or suffix that was matched by the
3185 router that accepted the address. It now behaves in the same way as it
3186 does for delivery: the affixes are removed from the local part unless
3187 rcpt_include_affixes is set on the transport.
3189 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3190 timeout during the DATA phase of an incoming message.
3192 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3193 applied a patch sanctioned by the Sieve maintainer.
3195 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3196 the uid or gid is negative. A case of a negative gid caused this to be
3197 noticed. The fix allows for either to be negative.
3199 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3200 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3203 PH/36 The change PH/12 above was broken. Fixed it.
3205 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3206 the grounds that routing the same address twice would always produce the
3207 same answer. This might have been true once, but it is certainly no
3208 longer true now. Routing a child address may depend on the previous
3209 routing that produced that child. Some complicated redirection strategies
3210 went wrong when messages had multiple recipients, and made Exim's
3211 behaviour dependent on the order in which the addresses were given.
3213 I have moved the duplicate checking until after the routing is complete.
3214 Exim scans the addresses that are assigned to local and remote
3215 transports, and removes any duplicates. This means that more work will be
3216 done, as duplicates will always all be routed, but duplicates are
3217 presumably rare, so I don't expect this is of any significance.
3219 For deliveries to pipes, files, and autoreplies, the duplicate checking
3220 still happens during the routing process, since they are not going to be
3223 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3224 It corrects a timeout issue with spamd. This is Ian's comment: "The
3225 background is that sometimes spamd either never reads data from a
3226 connection it has accepted, or it never writes response data. The exiscan
3227 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3228 blindly assumes that writes won't block so it may never time out."
3230 PH/39 Allow G after quota size as well as K and M.
3232 PH/40 The value set for $authenticated_id in an authenticator may not contain
3233 binary zeroes or newlines because the value is written to log lines and
3234 to spool files. There was no check on this. Now the value is run through
3235 the string_printing() function so that such characters are converted to
3236 printable escape sequences.
3238 PH/41 $message_linecount is a new variable that contains the total number of
3239 lines in the message. Compare $body_linecount, which is the count for the
3242 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3243 bounce and delay warning messages, except in certain special cases, which
3246 (a) An SMTP error message from a remote host;
3247 (b) A message specified in a :fail: redirection;
3248 (c) A message specified in a "fail" command in a system filter;
3249 (d) A message specified in a FAIL return from the queryprogram router;
3250 (e) A message specified by the cannot_route_message router option.
3252 In these cases only, Exim does include the error details in bounce and
3253 warning messages. There are also a few cases where bland messages such
3254 as "unrouteable address" or "local delivery error" are given.
3256 PH/43 $value is now also set for the "else" part of a ${run expansion.
3258 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3259 being worked on, but at least Exim now implements the latest version to
3262 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3263 process to complete was treated as a delivery failure, a timeout while
3264 writing the message to the pipe was logged, but erroneously treated as a
3265 successful delivery. Such timeouts include transport filter timeouts. For
3266 consistency with the overall process timeout, these timeouts are now
3267 treated as errors, giving rise to delivery failures by default. However,
3268 there is now a new Boolean option for the pipe transport called
3269 timeout_defer, which, if set TRUE, converts the failures into defers for
3270 both kinds of timeout. A transport filter timeout is now identified in
3273 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3274 systems where "make" and "gmake" are different, calling "gmake" at top
3275 level broke things. I've arranged for the value of $(MAKE) to be passed
3276 from the Makefile to this script so that it can call the same version of
3280 A note about Exim versions 4.44 and 4.50
3281 ----------------------------------------
3283 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3284 changes of various kinds. As a consequence, a big documentation update was
3285 needed. This delayed the release for rather longer than seemed good, especially
3286 in the light of a couple of (minor) security issues. Therefore, the changes
3287 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3288 release. So 4.44 and 4.50 are in effect two different branches that both start
3291 I have left the 4.50 change log unchanged; it contains all the changes since
3292 4.43. The change log for 4.44 is below; many of its items are identical to
3293 those for 4.50. This seems to be the most sensible way to preserve the
3294 historical information.
3300 1. Minor wording change to the doc/README.SIEVE file.
3302 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3303 computation of the current number of files was incorrect.
3305 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3306 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3309 4. Give more explanation in the error message when the command for a transport
3310 filter fails to execute.
3312 5. There are several places where Exim runs a non-Exim command in a
3313 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3314 was being done only for the command run by the queryprogram router. It is
3315 now done for all such subprocesses. The other cases are: ${run, transport
3316 filters, and the commands run by the lmtp and pipe transports.
3318 6. Added CONFIGURE_GROUP build-time option.
3320 7. Some older OS have a limit of 256 on the maximum number of file
3321 descriptors. Exim was using setrlimit() to set 1000 as a large value
3322 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3323 systems. I've change it so that if it can't get 1000, it tries for 256.
3325 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3326 was an oversight, and furthermore, ever since the addition of extra
3327 controls (e.g. 4.43/32), the checks on when to allow different forms of
3328 "control" were broken. There should now be diagnostics for all cases when a
3329 control that does not make sense is encountered.
3331 9. Added the /retain_sender option to "control=submission".
3333 10. $recipients is now available in the predata ACL (oversight).
3335 11. Tidy the search cache before the fork to do a delivery from a message
3336 received from the command line. Otherwise the child will trigger a lookup
3337 failure and thereby defer the delivery if it tries to use (for example) a
3338 cached ldap connection that the parent has called unbind on.
3340 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3341 of $address_data from the recipient verification was clobbered by the
3342 sender verification.
3344 13. The value of address_data from a sender verification is now available in
3345 $sender_address_data in subsequent conditions in the ACL statement.
3347 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3349 15. Added a new option "connect=<time>" to callout options, to set a different
3352 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3353 was its contents. (It was OK if the option was not defined at all.)
3355 17. A "Completed" log line is now written for messages that are removed from
3356 the spool by the -Mrm option.
3358 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3359 information about exactly what failed.
3361 19. Added -dd to debug only the daemon process.
3363 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3364 handles timeouts, both on the server side and network timeouts. Renamed the
3365 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3367 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3368 It is now set to "smtps".
3370 22. $host_address is now set to the target address during the checking of
3371 ignore_target_hosts.
3373 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3374 being passed; this would have caused $sender_host_name to have been used if
3375 matching the list had actually called for a host name (not very likely,
3376 since this list is usually IP addresses). A host name is now passed as
3379 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3380 code that specifies a non-restarting handler (typically sigaction() in
3381 modern systems) in an attempt to fix a rare and obscure crash bug.
3383 25. Narrowed the window for a race in the daemon that could cause it to ignore
3384 SIGCHLD signals. This is not a major problem, because they are used only to
3385 wake it up if nothing else does.
3387 26. A malformed maildirsize file could cause Exim to calculate negative values
3388 for the mailbox size or file count. Odd effects could occur as a result.
3389 The maildirsize information is now recalculated if the size or filecount
3392 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3393 support for a long time. Removed HAVE_SYS_VFS_H.
3395 28. Installed the latest version of exipick from John Jetmore.
3397 29. In an address list, if the pattern was not a regular expression, an empty
3398 subject address (from a bounce message) matched only if the pattern was an
3399 empty string. Non-empty patterns were not even tested. This was the wrong
3400 because it is perfectly reasonable to use an empty address as part of a
3401 database query. An empty address is now tested by patterns that are
3402 lookups. However, all the other forms of pattern expect the subject to
3403 contain a local part and a domain, and therefore, for them, an empty
3404 address still always fails if the pattern is not itself empty.
3406 30. Exim went into a mad DNS loop when attempting to do a callout where the
3407 host was specified on an smtp transport, and looking it up yielded more
3408 than one IP address.
3410 31. Re-factored the code for checking spool and log partition space into a
3411 function that finds that data and another that does the check. The former
3412 is then used to implement four new variables: $spool_space, $log_space,
3413 $spool_inodes, and $log_inodes.
3415 32. The RFC2047 encoding function was originally intended for short strings
3416 such as real names; it was not keeping to the 75-character limit for
3417 encoded words that the RFC imposes. It now respects the limit, and
3418 generates multiple encoded words if necessary. To be on the safe side, I
3419 have increased the buffer size for the ${rfc2047: expansion operator from
3422 33. It is now permitted to omit both strings after an "if" condition; if the
3423 condition is true, the result is "true". As before, when the second string
3424 is omitted, a false condition yields an empty string. This makes it less
3425 cumbersome to write custom ACL and router conditions.
3427 34. Failure to deliver a bounce message always caused it to be frozen, even if
3428 there was an errors_to setting on the router. The errors_to setting is now
3431 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3432 canonical form (fully expanded) before being placed in
3433 $sender_host_address.
3435 36. The table in the code that translates DNS record types into text (T_A to
3436 "A" for instance) was missing entries for NS and CNAME. It is just possible
3437 that this could have caused confusion if both these types were looked up
3438 for the same domain, because the text type is used as part of Exim's
3439 per-process caching. But the chance of anyone hitting this buglet seems
3442 37. The dnsdb lookup has been extended in a number of ways.
3444 (1) There is a new type, "zns", which walks up the domain tree until it
3445 finds some nameserver records. It should be used with care.
3447 (2) There is a new type, "mxh", which is like "mx" except that it returns
3448 just the host names, not the priorities.
3450 (3) It is now possible to give a list of domains (or IP addresses) to be
3451 looked up. The behaviour when one of the lookups defers can be
3452 controlled by a keyword.
3454 (4) It is now possible to specify the separator character for use when
3455 multiple records are returned.
3457 38. The dnslists ACL condition has been extended: it is now possible to supply
3458 a list of IP addresses and/or domains to be looked up in a particular DNS
3461 39. Added log_selector=+queue_time_overall.
3463 40. When running the queue in the test harness, wait just a tad after forking a
3464 delivery process, to get repeatability of debugging output.
3466 41. Include certificate and key file names in error message when GnuTLS fails
3467 to set them up, because the GnuTLS error message doesn't include the name
3468 of the failing file when there is a problem reading it.
3470 42. Allow both -bf and -bF in the same test run.
3472 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
3474 44. The "Exiscan patch" is now merged into the mainline Exim source.
3476 45. Sometimes the final signoff response after QUIT could fail to get
3477 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3478 < 0 before doing a fflush(). This bug looks as though it goes back to the
3479 introduction of TLS in release 3.20, but "sometimes" must have been rare
3480 because the tests only now provoked it.
3482 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
3483 (this can affect the format of dates).
3485 47. exim_tidydb, when checking for the continued existence of a message for
3486 which it has found a message-specific retry record, was not finding
3487 messages that were in split spool directories. Consequently, it was
3488 deleting retry records that should have stayed in existence.
3490 48. Steve fixed some bugs in eximstats.
3492 49. The SPA authentication driver was not abandoning authentication and moving
3493 on to the next authenticator when an expansion was forced to fail,
3494 contradicting the general specification for all authenticators. Instead it
3495 was generating a temporary error. It now behaves as specified.
3497 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
3498 (the order specifies the preference for clients). The order is now AES256,
3499 AES128, 3DES, ARCFOUR128.
3501 51. Small patch to Sieve code - explicitly set From: when generating an
3504 52. Exim crashed if a remote delivery caused a very long error message to be
3505 recorded - for instance if somebody sent an entire SpamAssassin report back
3506 as a large number of 550 error lines. This bug was coincidentally fixed by
3507 increasing the size of one of Exim's internal buffers (big_buffer) that
3508 happened as part of the Exiscan merge. However, to be on the safe side, I
3509 have made the code more robust (and fixed the comments that describe what
3512 53. Now that there can be additional text after "Completed" in log lines (if
3513 the queue_time_overall log selector is set), a one-byte patch to exigrep
3514 was needed to allow it to recognize "Completed" as not the last thing in
3517 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
3518 patch that reportedly fixes this has been added. I am not expert enough to
3519 create a test for it. This is what the patch creator wrote:
3521 "I found a little strange behaviour of ldap code when working with
3522 Windows 2003 AD Domain, where users was placed in more than one
3523 Organization Units. When I tried to give exim partial DN, the exit code
3524 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
3525 But simultaneously result of request was absolutely normal ldap result,
3526 so I produce this patch..."
3528 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
3529 REFERENCE, so I have modified the code to exclude the patch when that macro
3532 55. Some experimental protocols are using DNS PTR records for new purposes. The
3533 keys for these records are domain names, not reversed IP addresses. The
3534 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3535 leaves it alone. Component reversal etc. now happens only for IP addresses.
3538 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
3540 57. Double the size of the debug message buffer (to 2048) so that more of very
3541 long debug lines gets shown.
3543 58. The exicyclog utility now does better if the number of log files to keep
3544 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
3546 59. Two changes related to the smtp_active_hostname option:
3548 (1) $smtp_active_hostname is now available as a variable.
3549 (2) The default for smtp_banner uses $smtp_active_hostname instead
3550 of $primary_hostname.
3552 60. The host_aton() function is supposed to be passed a string that is known
3553 to be a valid IP address. However, in the case of IPv6 addresses, it was
3554 not checking this. This is a hostage to fortune. Exim now panics and dies
3555 if the condition is not met. A case was found where this could be provoked
3556 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3557 components; fortuitously, this particular loophole had already been fixed
3558 by change 4.50/55 above.
3560 If there are any other similar loopholes, the new check in host_aton()
3561 itself should stop them being exploited. The report I received stated that
3562 data on the command line could provoke the exploit when Exim was running as
3563 exim, but did not say which command line option was involved. All I could
3564 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3565 running as the user.
3568 61. There was a buffer overflow vulnerability in the SPA authentication code
3569 (which came originally from the Samba project). I have added a test to the
3570 spa_base64_to_bits() function which I hope fixes it.
3573 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
3574 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
3576 63. The daemon start-up calls getloadavg() while still root for those OS that
3577 need the first call to be done as root, but it missed one case: when
3578 deliver_queue_load_max is set with deliver_drop_privilege. This is
3579 necessary for the benefit of the queue runner, because there is no re-exec
3580 when deliver_drop_privilege is set.
3582 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
3583 This has been fixed.
3585 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3586 was in use, was not putting the data itself into the right store pool;
3587 consequently, it could be overwritten for a subsequent message in the same
3588 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3591 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
3593 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
3594 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
3595 the function were treating the return as a boolean value, which happened to
3596 work because 0=false and not-0=true, but is not correct code.
3598 68. The host_aton() function was not handling scoped IPv6 addresses (those
3599 with, for example, "%eth0" on the end) correctly.
3601 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
3602 negated items (that is, ~something) in unsigned ints. Some compilers
3603 apparently mutter when there is no cast.
3605 70. If an address verification called from an ACL failed, and did not produce a
3606 user-specific message (i.e. there was only a "system" message), nothing was
3607 put in $acl_verify_message. In this situation, it now puts the system
3610 71. Change 4.23/11 added synchronization checking at the start of an SMTP
3611 session; change 4.31/43 added the unwanted input to the log line - except
3612 that it did not do this in the start of session case. It now does.
3614 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
3615 This is wrong and can cause the other end to generate a synchronization
3616 error if it is another Exim or anything else that does the synchronization
3617 check. A QUIT command is no longer sent after a timeout.
3619 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
3620 during host lookups.
3622 74. The defer_ok option of callout verification was not working if it was used
3623 when verifying addresses in header lines, that is, for this case:
3625 verify = header_sender/callout=defer_ok
3627 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
3628 those file descriptors could be used for SMTP connections. If anything
3629 wrote to stderr (the example that came up was "warn" in embedded Perl), it
3630 could be sent to the SMTP client, causing chaos. The daemon now opens
3631 stdin, stdout, and stderr to /dev/null when it puts itself into the
3634 76. Arrange for output from Perl's "warn" command to be written to Exim's main
3635 log by default. The user can override this with suitable Perl magic.
3637 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
3638 the log message when discard triggers, was not working for the DATA ACL or
3639 for the non-SMTP ACL.
3641 78. Error message wording change in sieve.c.
3643 79. If smtp_accept_max_per_host was set, the number of connections could be
3644 restricted to fewer than expected, because the daemon was trying to set up
3645 a new connection before checking whether the processes handling previous
3646 connections had finished. The check for completed processes is now done
3647 earlier. On busy systems, this bug wouldn't be noticed because something
3648 else would have woken the daemon, and it would have reaped the completed
3651 80. If a message was submitted locally by a user whose login name contained one
3652 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
3653 It caused a spool format error. I have fixed the spool reading code. A
3654 related problem was that the "from" clause in the Received: line became
3655 illegal because of the space(s). It is now covered by ${quote_local_part.
3657 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
3660 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
3662 83. Patch from Sieve maintainer to fix unterminated string problem in
3663 "vacation" handling.
3665 84. Some minor changes to the Linux configuration files to help with other
3666 OS variants using glibc.
3668 85. One more patch for Sieve to update vacation handling to latest spec.
3671 ----------------------------------------------------
3672 See the note above about the 4.44 and 4.50 releases.
3673 ----------------------------------------------------
3679 1. Change 4.43/35 introduced a bug that caused file counts to be
3680 incorrectly computed when quota_filecount was set in an appendfile
3683 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3684 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3687 3. Give more explanation in the error message when the command for a transport
3688 filter fails to execute.
3690 4. There are several places where Exim runs a non-Exim command in a
3691 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3692 was being done only for the command run by the queryprogram router. It is
3693 now done for all such subprocesses. The other cases are: ${run, transport
3694 filters, and the commands run by the lmtp and pipe transports.
3696 5. Some older OS have a limit of 256 on the maximum number of file
3697 descriptors. Exim was using setrlimit() to set 1000 as a large value
3698 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3699 systems. I've change it so that if it can't get 1000, it tries for 256.
3701 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3702 was an oversight, and furthermore, ever since the addition of extra
3703 controls (e.g. 4.43/32), the checks on when to allow different forms of
3704 "control" were broken. There should now be diagnostics for all cases when a
3705 control that does not make sense is encountered.
3707 7. $recipients is now available in the predata ACL (oversight).
3709 8. Tidy the search cache before the fork to do a delivery from a message
3710 received from the command line. Otherwise the child will trigger a lookup
3711 failure and thereby defer the delivery if it tries to use (for example) a
3712 cached ldap connection that the parent has called unbind on.
3714 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3715 of $address_data from the recipient verification was clobbered by the
3716 sender verification.
3718 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3719 was its contents. (It was OK if the option was not defined at all.)
3721 11. A "Completed" log line is now written for messages that are removed from
3722 the spool by the -Mrm option.
3724 12. $host_address is now set to the target address during the checking of
3725 ignore_target_hosts.
3727 13. When checking ignore_target_hosts for an ipliteral router, no host name was
3728 being passed; this would have caused $sender_host_name to have been used if
3729 matching the list had actually called for a host name (not very likely,
3730 since this list is usually IP addresses). A host name is now passed as
3733 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3734 code that specifies a non-restarting handler (typically sigaction() in
3735 modern systems) in an attempt to fix a rare and obscure crash bug.
3737 15. Narrowed the window for a race in the daemon that could cause it to ignore
3738 SIGCHLD signals. This is not a major problem, because they are used only to
3739 wake it up if nothing else does.
3741 16. A malformed maildirsize file could cause Exim to calculate negative values
3742 for the mailbox size or file count. Odd effects could occur as a result.
3743 The maildirsize information is now recalculated if the size or filecount
3746 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3747 support for a long time. Removed HAVE_SYS_VFS_H.
3749 18. Updated exipick to current release from John Jetmore.
3751 19. Allow an empty sender to be matched against a lookup in an address list.
3752 Previously the only cases considered were a regular expression, or an
3755 20. Exim went into a mad DNS lookup loop when doing a callout where the
3756 host was specified on the transport, if the DNS lookup yielded more than
3759 21. The RFC2047 encoding function was originally intended for short strings
3760 such as real names; it was not keeping to the 75-character limit for
3761 encoded words that the RFC imposes. It now respects the limit, and
3762 generates multiple encoded words if necessary. To be on the safe side, I
3763 have increased the buffer size for the ${rfc2047: expansion operator from
3766 22. Failure to deliver a bounce message always caused it to be frozen, even if
3767 there was an errors_to setting on the router. The errors_to setting is now
3770 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3771 canonical form (fully expanded) before being placed in
3772 $sender_host_address.
3774 24. Updated eximstats to version 1.33
3776 25. Include certificate and key file names in error message when GnuTLS fails
3777 to set them up, because the GnuTLS error message doesn't include the name
3778 of the failing file when there is a problem reading it.
3780 26. Expand error message when OpenSSL has problems setting up cert/key files.
3783 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
3784 (this can affect the format of dates).
3786 28. exim_tidydb, when checking for the continued existence of a message for
3787 which it has found a message-specific retry record, was not finding
3788 messages that were in split spool directories. Consequently, it was
3789 deleting retry records that should have stayed in existence.
3791 29. eximstats updated to version 1.35
3792 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
3793 1.35 - bugfix such that pie charts by volume are generated correctly
3795 30. The SPA authentication driver was not abandoning authentication and moving
3796 on to the next authenticator when an expansion was forced to fail,
3797 contradicting the general specification for all authenticators. Instead it
3798 was generating a temporary error. It now behaves as specified.
3800 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
3801 (the order specifies the preference for clients). The order is now AES256,
3802 AES128, 3DES, ARCFOUR128.
3804 31. Small patch to Sieve code - explicitly set From: when generating an
3807 32. Exim crashed if a remote delivery caused a very long error message to be
3808 recorded - for instance if somebody sent an entire SpamAssassin report back
3809 as a large number of 550 error lines. This bug was coincidentally fixed by
3810 increasing the size of one of Exim's internal buffers (big_buffer) that
3811 happened as part of the Exiscan merge. However, to be on the safe side, I
3812 have made the code more robust (and fixed the comments that describe what
3815 33. Some experimental protocols are using DNS PTR records for new purposes. The
3816 keys for these records are domain names, not reversed IP addresses. The
3817 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3818 leaves it alone. Component reversal etc. now happens only for IP addresses.
3821 34. The host_aton() function is supposed to be passed a string that is known
3822 to be a valid IP address. However, in the case of IPv6 addresses, it was
3823 not checking this. This is a hostage to fortune. Exim now panics and dies
3824 if the condition is not met. A case was found where this could be provoked
3825 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3826 components; fortuitously, this particular loophole had already been fixed
3827 by change 4.50/55 or 4.44/33 above.
3829 If there are any other similar loopholes, the new check in host_aton()
3830 itself should stop them being exploited. The report I received stated that
3831 data on the command line could provoke the exploit when Exim was running as
3832 exim, but did not say which command line option was involved. All I could
3833 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3834 running as the user.
3837 35. There was a buffer overflow vulnerability in the SPA authentication code
3838 (which came originally from the Samba project). I have added a test to the
3839 spa_base64_to_bits() function which I hope fixes it.
3842 36. The daemon start-up calls getloadavg() while still root for those OS that
3843 need the first call to be done as root, but it missed one case: when
3844 deliver_queue_load_max is set with deliver_drop_privilege. This is
3845 necessary for the benefit of the queue runner, because there is no re-exec
3846 when deliver_drop_privilege is set.
3848 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3849 was in use, was not putting the data itself into the right store pool;
3850 consequently, it could be overwritten for a subsequent message in the same
3851 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3854 38. Sometimes the final signoff response after QUIT could fail to get
3855 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3856 < 0 before doing a fflush(). This bug looks as though it goes back to the
3857 introduction of TLS in release 3.20, but "sometimes" must have been rare
3858 because the tests only now provoked it.
3864 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
3865 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
3866 Now it returns an integer. A number of calls were still expecting a T/F
3867 return. Fortuitously, in all cases, the tests worked in OK situations,
3868 which is the norm. However, things would have gone wrong on any write
3869 failures on the smtp file descriptor. This function is used when sending
3870 messages over SMTP and also when doing verify callouts.
3872 2. When Exim is called to do synchronous delivery of a locally submitted
3873 message (the -odf or -odi options), it no longer closes stderr before doing
3876 3. Implemented the mua_wrapper option.
3878 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
3880 5. Implemented the functions header_remove(), header_testname(),
3881 header_add_at_position(), and receive_remove_recipient(), and exported them
3884 6. If an ACL "warn" statement specified the addition of headers, Exim already
3885 inserted X-ACL-Warn: at the start if there was no header name. However, it
3886 was not making this test for the second and subsequent header lines if
3887 there were newlines in the string. This meant that an invalid header could
3888 be inserted if Exim was badly configured.
3890 7. Allow an ACL "warn" statement to add header lines at the start or after all
3891 the Received: headers, as well as at the end.
3893 8. Added the rcpt_4xx retry error code.
3895 9. Added postmaster_mailfrom=xxx to callout verification option.
3897 10. Added mailfrom=xxxx to the callout verification option, for verify=
3900 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
3901 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
3903 12. Inserted some casts to stop certain compilers warning when using pointer
3904 differences as field lengths or precisions in printf-type calls (mostly
3905 affecting debugging statements).
3907 13. Added optional readline() support for -be (dynamically loaded).
3909 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
3910 same clock tick as a message's arrival, so that its received time was the
3911 same as the "first fail" time on the retry record, and that message
3912 remained on the queue past the ultimate address timeout, every queue runner
3913 would try a delivery (because it was past the ultimate address timeout) but
3914 after another failure, the ultimate address timeout, which should have then
3915 bounced the address, did not kick in. This was a "< instead of <=" error;
3916 in most cases the first failure would have been in the next clock tick
3917 after the received time, and all would be well.
3919 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
3920 being recognized when the domain list was tested by the match_domain
3921 condition in an expansion string.
3923 16. Added the ${str2b64: operator.
3925 17. Exim was always calling setrlimit() to set a large limit for the number of
3926 processes, without checking whether the existing limit was already
3927 adequate. (It did check for the limit on file descriptors.) Furthermore,
3928 errors from getrlimit() and setrlimit() were being ignored. Now they are
3929 logged to the main and panic logs, but Exim does carry on, to try to do its
3930 job under whatever limits there are.
3932 18. Imported PCRE 5.0.
3934 19. Trivial typo in log message " temporarily refused connection" (the leading
3937 20. If the log selector return_path_on_delivery was set and an address was
3938 redirected to /dev/null, the delivery process crashed because it assumed
3939 that a return path would always be set for a "successful" delivery. In this
3940 case, the whole delivery is bypassed as an optimization, and therefore no
3943 21. Internal re-arrangement: the function for sending a challenge and reading
3944 a response while authentication was assuming a zero-terminated challenge
3945 string. It's now changed to take a pointer and a length, to allow for
3946 binary data in such strings.
3948 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3950 23. Exim was not respecting finduser_retries when seeking the login of the
3951 uid under which it was called; it was always trying 10 times. (The default
3952 setting of finduser_retries is zero.) Also, it was sleeping after the final
3953 failure, which is pointless.
3955 24. Implemented tls_on_connect_ports.
3957 25. Implemented acl_smtp_predata.
3959 26. If the domain in control=submission is set empty, Exim assumes that the
3960 authenticated id is a complete email address when it generates From: or
3961 Sender: header lines.
3963 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3964 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3965 chown and chgrp in /bin and hostname in /usr/bin.
3967 28. Exim was keeping the "process log" file open after each use, just as it
3968 does for the main log. This opens the possibility of it remaining open for
3969 long periods when the USR1 signal hits a daemon. Occasional processlog
3970 errors were reported, that could have been caused by this. Anyway, it seems
3971 much more sensible not to leave this file open at all, so that is what now
3974 29. The long-running daemon process does not normally write to the log once it
3975 has entered its main loop, and it closes the log before doing so. This is
3976 so that log files can straightforwardly be renamed and moved. However,
3977 there are a couple of unusual error situations where the daemon does write
3978 log entries, and I had neglected to close the log afterwards.
3980 30. The text of an SMTP error response that was received during a remote
3981 delivery was being truncated at 512 bytes. This is too short for some of
3982 the long messages that one sometimes sees. I've increased the limit to
3985 31. It is now possible to make retry rules that apply only when a message has a
3986 specific sender, in particular, an empty sender.
3988 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3989 it possible to be selective about when SMTP synchronization is enforced.
3991 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3993 32. Implemented hosts_connection_nolog.
3995 33. Added an ACL for QUIT.
3997 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4000 35. Added mailbox_size and mailbox_filecount to appendfile.
4002 36. Added control = no_multiline_responses to ACLs.
4004 37. There was a bug in the logic of the code that waits for the clock to tick
4005 in the case where the clock went backwards by a substantial amount such
4006 that the microsecond fraction of "now" was more than the microsecond
4007 fraction of "then" (but the whole seconds number was less).
4009 38. Added support for the libradius Radius client library this is found on
4010 FreeBSD (previously only the radiusclient library was supported).
4016 1. When certain lookups returned multiple values in the form name=value, the
4017 quoting of the values was not always being done properly. Specifically:
4018 (a) If the value started with a double quote, but contained no whitespace,
4020 (b) If the value contained whitespace other than a space character (i.e.
4021 tabs or newlines or carriage returns) it was not quoted.
4022 This fix has been applied to the mysql and pgsql lookups by writing a
4023 separate quoting function and calling it from the lookup code. The fix
4024 should probably also be applied to nisplus, ibase and oracle lookups, but
4025 since I cannot test any of those, I have not disturbed their existing code.
4027 2. A hit in the callout cache for a specific address caused a log line with no
4028 reason for rejecting RCPT. Now it says "Previous (cached) callout
4029 verification failure".
4031 3. There was an off-by-one bug in the queryprogram router. An over-long
4032 return line was truncated at 256 instead of 255 characters, thereby
4033 overflowing its buffer with the terminating zero. As well as fixing this, I
4034 have increased the buffer size to 1024 (and made a note to document this).
4036 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4037 when Exim is waiting for an SMTP response from a remote server, Exim
4038 restarts its select() call on the socket, thereby resetting its timeout.
4039 This is not a problem when such interrupts are rare. Somebody set up a cron
4040 job to run exiwhat every 2 minutes, which is less than the normal select()
4041 timeout (5 or 10 minutes). This meant that the select() timeout never
4042 kicked in because it was always reset. I have fixed this by comparing the
4043 time when an interrupt arrives with the time at the start of the first call
4044 to select(). If more time than the timeout has elapsed, the interrupt is
4045 treated as a timeout.
4047 5. Some internal re-factoring in preparation for the addition of Sieve
4048 extensions (by MH). In particular, the "personal" test is moved to a
4049 separate function, and given an option for scanning Cc: and Bcc: (which is
4050 not set for Exim filters).
4052 6. When Exim created an email address using the login of the caller as the
4053 local part (e.g. when creating a From: or Sender: header line), it was not
4054 quoting the local part when it contained special characters such as @.
4056 7. Installed new OpenBSD configuration files.
4058 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4059 try to make them clearer.
4061 9. Callout options, other than the timeout value, were being ignored when
4062 verifying sender addresses in header lines. For example, when using
4064 verify = header_sender/callout=no_cache
4066 the cache was (incorrectly) being used.
4068 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4069 only the Cygwin environment.
4071 11. When return_path_on_delivery was set as a log selector, if different remote
4072 addresses in the same message used different return paths and parallel
4073 remote delivery occurred, the wrong values would sometimes be logged.
4074 (Whenever a remote delivery process finished, the return path value from
4075 the most recently started remote delivery process was logged.)
4077 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4078 lines when AUTH and/or TLS are in use. This is the "received protocol"
4079 field. Exim used to use "asmtp" for authenticated SMTP, without any
4080 indication (in the protocol name) for TLS use. Now it follows the RFC and
4081 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4082 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4083 names appear in log lines as well as in Received: header lines.
4085 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4086 extensions, and comparison tests, and to fix some bugs.
4088 14. Changes to the "personal" filter test:
4090 (1) The test was buggy in that it was just doing the equivalent of
4091 "contains" tests on header lines. For example, if a user's address was
4092 anne@some.where, the "personal" test would incorrectly be true for
4094 To: susanne@some.where
4096 This test is now done by extracting each address from the header in turn,
4097 and checking the entire address. Other tests that are part of "personal"
4098 are now done using regular expressions (for example, to check local parts
4099 of addresses in From: header lines).
4101 (2) The list of non-personal local parts in From: addresses has been
4102 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4103 taken from the Sieve specification recommendations.
4105 (3) If the message contains any header line starting with "List-" it is
4106 treated as non-personal.
4108 (4) The test for "circular" in the Subject: header line has been removed
4109 because it now seems ill-conceived.
4111 15. Minor typos in src/EDITME comments corrected.
4113 16. Installed latest exipick from John Jetmore.
4115 17. If headers_add on a router specified a text string that was too long for
4116 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4117 of string_sprintf() is now avoided.
4119 18. $message_body_size was not set (it was always zero) when running the DATA
4120 ACL and the local_scan() function.
4122 19. For the "mail" command in an Exim filter, no default was being set for
4123 the once_repeat time, causing a random time value to be used if "once" was
4124 specified. (If the value happened to be <= 0, no repeat happened.) The
4125 default is now 0s, meaning "never repeat". The "vacation" command was OK
4126 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4127 (I found it when inspecting the code).
4129 20. There is now an overall timeout for performing a callout verification. It
4130 defaults to 4 times the callout timeout, which applies to individual SMTP
4131 commands during the callout. The overall timeout applies when there is more
4132 than one host that can be tried. The timeout is checked before trying the
4133 next host. This prevents very long delays if there are a large number of
4134 hosts and all are timing out (e.g. when the network connections are timing
4135 out). The value of the overall timeout can be changed by specifying an
4136 additional sub-option for "callout", called "maxwait". For example:
4138 verify = sender/callout=5s,maxwait=20s
4140 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4141 to the end before writing, but this should make it even safer).
4143 22. Exim was forgetting that it had advertised PIPELINING for the second and
4144 subsequent messages on an SMTP connection. It was also not resetting its
4145 memory on STARTTLS and an internal HELO.
4147 23. When Exim logs an SMTP synchronization error within a session, it now
4148 records whether PIPELINING has been advertised or not.
4150 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4151 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4152 rather than long int.
4154 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4156 26. Added the never_mail option to autoreply.
4162 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4163 crash if the getsockname() call failed; this can happen if a connection is
4164 closed very soon after it is established. The problem was simply in the
4165 order in which certain operations were done, causing Exim to try to write
4166 to the SMTP stream before it had set up the file descriptor. The bug has
4167 been fixed by making things happen in the correct order.
4173 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4174 before the connection was closed, thus losing the rejection response.
4176 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4177 some early Solaris releases, but causes trouble in current releases where
4178 socklen_t is defined.
4180 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4183 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4186 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4187 that releasing the top of it at the end releases what was used for sub-
4188 expansions (unless the block got too big). However, discard this block if
4189 the first thing is a variable or header, so that we can use its block when
4190 it is dynamic (useful for very large $message_headers, for example).
4192 6. Lookups now cache *every* query, not just the most recent. A new, separate
4193 store pool is used for this. It can be recovered when all lookup caches are
4194 flushed. Lookups now release memory at the end of their result strings.
4195 This has involved some general refactoring of the lookup sources.
4197 7. Some code has been added to the store_xxx() functions to reduce the amount
4198 of flapping under certain conditions.
4200 8. log_incoming_interface used to affect only the <= reception log lines. Now
4201 it causes the local interface and port to be added to several more SMTP log
4202 lines, for example "SMTP connection from", and rejection lines.
4204 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4206 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4208 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4209 could be overwritten at the end of the current message (or the start of a
4210 new message if it was set in a HELO ACL). The value is now preserved for
4211 the duration of the SMTP connection.
4213 12. If a transport had a headers_rewrite setting, and a matching header line
4214 contained an unqualified address, that address was qualified, even if it
4215 did not match any rewriting rules. The underlying bug was that the values
4216 of the flags that permit the existence of unqualified sender and recipient
4217 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4218 non-local messages, and by -bnq for local messages) were not being
4219 preserved with the message after it was received.
4221 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4222 "next input=" as part of the text comprising the host identity instead of
4223 the correct text. The code was using the same buffer for two different
4224 strings. However, depending on which order the printing function evaluated
4225 its arguments, the bug did not always show up. Under Linux, for example, my
4226 test suite worked just fine.
4228 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4229 doesn't work with some older versions of Perl. It has been changed to "my",
4230 which in any case is probably the better facility to use.
4232 15. A really picky compiler found some instances of statements for creating
4233 error messages that either had too many or two few arguments for the format
4236 16. The size of the buffer for calls to the DNS resolver has been increased
4237 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4238 for addresses that have a lot of PTR records. This alleviates a problem; it
4239 does not fully solve it.
4241 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4242 buffer now truncates the list and logs the incident, which is the same
4243 action as happens when Exim is looking up a host name and its aliases.
4244 Previously in this situation something unpredictable would happen;
4245 sometimes it was "internal error: store_reset failed".
4247 18. If a server dropped the connection unexpectedly when an Exim client was
4248 using GnuTLS and trying to read a response, the client delivery process
4249 crashed while trying to generate an error log message.
4251 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4252 string, for example:
4254 warn message = H1: something\nH2: something
4256 the text was added as a single header line from Exim's point of view
4257 though it ended up OK in the delivered message. However, searching for the
4258 second and subsequent header lines using $h_h2: did not work. This has been
4259 fixed. Similarly, if a system filter added multiple headers in this way,
4260 the routers could not see them.
4262 20. Expanded the error message when iplsearch is called with an invalid key to
4263 suggest using net-iplsearch in a host list.
4265 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4266 ACLs are no longer actually imposed (and a message to that effect is
4269 22. If a "gecos" field in a passwd entry contained escaped characters, in
4270 particular, if it contained a \" sequence, Exim got it wrong when building
4271 a From: or a Sender: header from that name. A second bug also caused
4272 incorrect handling when an unquoted " was present following a character
4273 that needed quoting.
4275 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4276 was not being matched caselessly.
4278 24. Arranged for all hyphens in the exim.8 source to be escaped with
4281 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4282 properly. Recipient callout cache records were still being keyed to include
4283 the sender, even when use_sender was set false. This led to far more
4284 callouts that were necessary. The sender is no longer included in the key
4285 when use_sender is false.
4287 26. Added "control = submission" modifier to ACLs.
4289 27. Added the ${base62d: operator to decode base 62 numbers.
4291 28. dnsdb lookups can now access SRV records.
4293 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4294 the configuration file.
4296 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4297 (-v) mode. This makes the output for a verbose queue run more intelligible.
4299 31. Added a use_postmaster feature to recipient callouts.
4301 32. Added the $body_zerocount variable, containing the number of binary zero
4302 bytes in the message body.
4304 33. The time of last modification of the "new" subdirectory is now used as the
4305 "mailbox time last read" when there is a quota error for a maildir
4308 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4310 35. Added +ignore_unknown as a special item in host lists.
4312 36. Code for decoding IPv6 addresses in host lists is now included, even if
4313 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4314 address was recognized as an IP address, but was then not correctly decoded
4315 into binary, causing unexpected and incorrect effects when compared with
4322 1. Very minor rewording of debugging text in manualroute to say "list of
4323 hosts" instead of "hostlist".
4325 2. If verify=header_syntax was set, and a header line with an unqualified
4326 address (no domain) and a large number of spaces between the end of the
4327 name and the colon was received, the reception process suffered a buffer
4328 overflow, and (when I tested it) crashed. This was caused by some obsolete
4329 code that should have been removed. The fix is to remove it!
4331 3. When running in the test harness, delay a bit after writing a bounce
4332 message to get a bit more predictability in the log output.
4334 4. Added a call to search_tidyup() just before forking a reception process. In
4335 theory, someone could use a lookup in the expansion of smtp_accept_max_
4336 per_host which, without the tidyup, could leave open a database connection.
4338 5. Added the variables $recipient_data and $sender_data which get set from a
4339 lookup success in an ACL "recipients" or "senders" condition, or a router
4340 "senders" option, similar to $domain_data and $local_part_data.
4342 6. Moved the writing of debug_print from before to after the "senders" test
4345 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4346 problems for message scanning, either using a data ACL, or using
4347 local_scan() because the Received: header was not generated till after they
4348 were called (in order to set the time as the time of reception completion).
4349 I have revised the way this works. The header is now generated after the
4350 body is received, but before the ACL or local_scan() are called. After they
4351 are run, the timestamp in the header is updated.
4357 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4358 before starting a queue runner without re-exec. This happened only when
4359 deliver_drop_privilege was set or when the Exim user was set to root. The
4360 effect of the bug was that timeouts during subsequent deliveries caused
4361 crashes instead of being properly handled. The handler is now left at its
4362 default (and expected) setting.
4364 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4365 message, again when deliver_drop_privilege is set or Exim is run as root.
4366 The bug described in (1) was not present in this case, but the tidying up
4367 of the other signals was missing. I have made the two cases consistent.
4369 3. The ignore_target_hosts setting on a manualroute router was being ignored
4370 for hosts that were looked up using the /MX notation.
4372 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4375 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4376 operated on the sender address. After changing the $sender_address to <>
4377 for the sender address verify, Exim was re-instated it as the original
4378 (before rewriting) address, but remembering that it had rewritten it, so it
4379 wasn't rewriting it again. This bug also had the effect of breaking the
4380 sender address verification caching when the sender address was rewritten.
4382 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4383 This has been changed so that if the ip literal address matches
4384 ignore_target_hosts, the router declines.
4386 7. Added expansion conditions match_domain, match_address, and match_local_
4387 part (NOT match_host).
4389 8. The placeholder for the Received: header didn't have a length field set.
4391 9. Added code to Exim itself and to exim_lock to test for a specific race
4392 condition that could lead to file corruption when using MBX delivery. The
4393 issue is with the lockfile that is created in /tmp. If this file is removed
4394 after a process has opened it but before that process has acquired a lock,
4395 there is the potential for a second process to recreate the file and also
4396 acquire a lock. This could lead to two Exim processes writing to the file
4397 at the same time. The added code performs the same test as UW imapd; it
4398 checks after acquiring the lock that its file descriptor still refers to
4399 the same named file.
4401 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4402 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4405 11. Added the smtp_active_hostname option. If used, this will typically be made
4406 to depend on the incoming interface address. Because $interface_address is
4407 not set up until the daemon has forked a reception process, error responses
4408 that can happen earlier (such as "too many connections") no longer contain
4411 12. If an expansion in a condition on a "warn" statement fails because a lookup
4412 defers, the "warn" statement is abandoned, and the next ACL statement is
4413 processed. Previously this caused the whole ACL to be aborted.
4415 13. Added the iplsearch lookup type.
4417 14. Added ident_timeout as a log selector.
4419 15. Added tls_certificate_verified as a log selector.
4421 16. Added a global option tls_require_ciphers (compare the smtp transport
4422 option of the same name). This controls incoming TLS connections.
4424 17. I finally figured out how to make tls_require_ciphers do a similar thing
4425 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
4426 before starting the TLS session.
4428 18. Tabs are now shown as \t in -bP output.
4430 19. If the log selector return_path_on_delivery was set, Exim crashed when
4431 bouncing a message because it had too many Received: header lines.
4433 20. If two routers both had headers_remove settings, and the first one included
4434 a superfluous trailing colon, the final name in the first list and the
4435 first name in the second list were incorrectly joined into one item (with a
4436 colon in the middle).
4442 1. Added -C and -D options to the exinext utility, mainly to make it easier
4443 to include in the automated testing, but these could be helpful when
4444 multiple configurations are in use.
4446 2. The exinext utility was not formatting the output nicely when there was
4447 an alternate port involved in the retry record key, nor when there was a
4448 message id as well (for retries that were specific to a specific message
4449 and a specific host). It was also confused by IPv6 addresses, because of
4450 the additional colons they contain. I have fixed the IPv4 problem, and
4451 patched it up to do a reasonable job for IPv6.
4453 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
4454 delivery, the log line now contains "pipelined" if PIPELINING was used.
4456 4. An SMTP transport process used to panic and die if the bind() call to set
4457 an explicit outgoing interface failed. This has been changed; it is now
4458 treated in the same way as a connect() failure.
4460 5. A reference to $sender_host_name in the part of a conditional expansion
4461 that was being skipped was still causing a DNS lookup. This no longer
4464 6. The def: expansion condition was not recognizing references to header lines
4465 that used bh_ and bheader_.
4467 7. Added the _cache feature to named lists.
4469 8. The code for checking quota_filecount in the appendfile transport was
4470 allowing one more file than it should have been.
4472 9. For compatibility with Sendmail, the command line option
4480 and sets the incoming protocol and host name (for trusted callers). The
4481 host name and its colon can be omitted when only the protocol is to be set.
4482 Note the Exim already has two private options, -pd and -ps, that refer to
4483 embedded Perl. It is therefore impossible to set a protocol value of "d" or
4484 "s", but I don't think that's a major issue.
4486 10. A number of refactoring changes to the code, none of which should affect
4489 (a) The number of logging options was getting close to filling up the
4490 32-bit word that was used as a bit map. I have split them into two classes:
4491 those that are passed in the argument to log_write(), and those that are
4492 only ever tested independently outside of that function. These are now in
4493 separate 32-bit words, so there is plenty of room for expansion again.
4494 There is no change in the user interface or the logging behaviour.
4496 (b) When building, for example, log lines, the code previously used a
4497 macro that called string_cat() twice, in order to add two strings. This is
4498 not really sufficiently general. Furthermore, there was one instance where
4499 it was actually wrong because one of the argument was used twice, and in
4500 one call a function was used. (As it happened, calling the function twice
4501 did not affect the overall behaviour.) The macro has been replaced by a
4502 function that can join an arbitrary number of extra strings onto a growing
4505 (c) The code for expansion conditions now uses a table and a binary chop
4506 instead of a serial search (which was left over from when there were very
4507 few conditions). Also, it now recognizes conditions like "pam" even when
4508 the relevant support is not compiled in: a suitably worded error message is
4509 given if an attempt is made to use such a condition.
4511 11. Added ${time_interval:xxxxx}.
4513 12. A bug was causing one of the ddress fields not to be passed back correctly
4514 from remote delivery subprocesses. The field in question was not being
4515 subsequently used, so this caused to problems in practice.
4517 13. Added new log selectors queue_time and deliver_time.
4519 14. Might have fixed a bug in maildirsizefile handling that threw up
4520 "unexpected character" debug warnings, and recalculated the data
4521 unnecessarily. In any case, I expanded the warning message to give more
4524 15. Added the message "Restricted characters in address" to the statements in
4525 the default ACL that block characters like @ and % in local parts.
4527 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
4528 Three changes have been made:
4530 (a) There was a serious bug; a negative response to MAIL caused the whole
4531 recipient domain to be cached as invalid, thereby blocking all messages
4532 to all local parts at the same domain, from all senders. This bug has
4533 been fixed. The domain is no longer cached after a negative response to
4534 MAIL if the sender used is not empty.
4536 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
4539 (c) A new callout option, "use_sender" has been added for people who want
4540 the modified behaviour.
4546 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
4549 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
4550 indeed breaks things for older releases.
4552 3. Added additional logging to the case where there is a problem reading data
4553 from a filter that is running in a subprocess using a pipe, in order to
4554 try to track down a specific problem.
4556 4. Testing facility fudge: when running in the test harness and attempting
4557 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
4558 getting "No route to host". Convert this to a timeout.
4560 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
4563 6. Some OS don't have socklen_t but use size_t instead. This affects the
4564 fifth argument of getsockopt() amongst other things. This is now
4565 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
4566 can be set for individual OS. I have set it for SunOS5, OSF1, and
4567 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
4568 some earlier ones do not.
4570 7. Change 4.30/15 was not doing the test caselessly.
4572 8. The standard form for an IPv6 address literal was being rejected by address
4573 parsing in, for example, MAIL and RCPT commands. An example of this kind of
4574 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4575 this, as well as the form without the "IPv6" on the front (but only when
4576 address literals are enabled, of course).
4578 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
4580 10. Exim crashed if a message with an empty sender address specified by -f
4581 encountered a router with an errors_to setting. This could be provoked only
4582 by a command such as
4586 where an empty string was supplied; "<>" did not hit this bug.
4588 11. Installed PCRE release 4.5.
4590 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
4591 remained set. It is now erased.
4593 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
4594 times from message ids (which are base 36 rather than the normal 62).
4596 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
4597 were being counted as actual protocol errors, and logged if the log
4598 selector +smtp_protocol_error was set. One cannot be perfect in this test,
4599 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
4600 and DATA following a set of rejected RCPTs do not count as protocol errors.
4601 In other words, Exim assumes they were pipelined, though this may not
4602 actually be the case. Of course, in all cases the client gets an
4603 appropriate error code.
4605 15. If a lookup fails in an ACL condition, a message about the failure may
4606 be available; it is used if testing the ACL cannot continue, because most
4607 such messages specify what the cause of the deferral is. However, some
4608 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
4609 that caused an old message to be retained and used if a later statement
4610 caused a defer, replacing the real cause of the deferral.
4612 16. If an IP address had so many PTR records that the DNS lookup buffer
4613 was not large enough to hold them, Exim could crash while trying to process
4614 the truncated data. It now detects and logs this case.
4616 17. Further to 4.21/58, another change has been made: if (and only if) the
4617 first line of a message (the first header line) ends with CRLF, a bare LF
4618 in a subsequent header line has a space inserted after it, so as not to
4619 terminate the header.
4621 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
4622 unnecessarily, used atoi() instead of strtol(), and didn't check the
4623 termination when getting file sizes from file names by regex.
4625 19. Completely re-implemented the support for maildirsize files, in the light
4626 of a number of problems with the previous contributed implementation
4627 (4.30/29). In particular:
4629 . If the quota is zero, the maildirsize file is maintained, but no quota is
4632 . If the maildir directory does not exist, it is created before any attempt
4633 to write a maildirsize file.
4635 . The quota value in the file is just a cache; if the quota is changed in
4636 the transport, the new value overrides.
4638 . A regular expression is available for excluding directories from the
4641 20. The autoreply transport checks the characters in options that define the
4642 message's headers; it allows continued headers, but it was checking with
4643 isspace() after an embedded newline instead of explicitly looking for a
4646 21. If all the "regular" hosts to which an address was routed had passed their
4647 expiry times, and had not reached their retry times, the address was
4648 bounced, even if fallback hosts were defined. Now Exim should go on to try
4651 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
4652 equivalent code in the SMTP transport. Some hosts send humungous responses
4653 to HELO/EHLO, more than 1024 it seems.
4655 23. Refactoring: code in filter.c used (void *) for "any old type" but this
4656 gives compiler warnings in some environments. I've now done it "properly",
4659 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
4660 (because of problems with the built-in one) was declared to return uschar *
4661 instead of char *, causing compiler failure.
4663 25. Fixed a file descriptor leak when processing alias/forward files.
4665 26. Fixed a minor format string issue in dbfn.c.
4667 27. Typo in exim.c: ("dmbnz" for "dbmnz").
4669 28. If a filter file refered to $h_xxx or $message_headers, and the headers
4670 contained RFC 2047 "words", Exim's memory could, under certain conditions,
4673 29. When a sender address is verified, it is cached, to save repeating the test
4674 when there is more than one recipient in a message. However, when the
4675 verification involves a callout, it is possible for different callout
4676 options to be set for different recipients. It is too complicated to keep
4677 track of this in the cache, so now Exim always runs a verification when a
4678 callout is required, relying on the callout cache for the optimization.
4679 The overhead is duplication of the address routing, but this should not be
4682 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
4683 to be verified with callout=postmaster, and the main callout worked but the
4684 postmaster check failed, the verification correctly failed. However, if a
4685 subsequent RCPT command asked for sender verification *without* the
4686 postmaster check, incorrect caching caused this verification also to fail,
4689 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
4690 it was not caching the DNS options (qualify_single, search_parents) that
4691 were used when the lookup failed. A subsequent lookup with different
4692 options therefore always gave the same answer, though there were cases
4693 where it should not have. (Example: a "domains = !$mx_any" option on a
4694 dnslookup router: the "domains" option is always processed without any
4695 widening, but the router might have qualify_single set.) Now Exim uses the
4696 cached value only when the same options are set.
4698 32. Added John Jetmore's "exipick" utility to the distribution.
4700 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
4701 than a timeout (e.g. a certificate is required, and is not provided), an
4702 Exim server now closes the connection immediately. Previously it waited for
4703 the client to close - but if the client is SSL, it seems that they each
4704 wait for each other, leading to a delay before one of them times out.
4706 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
4707 maintained 0.8.x compatibility because I don't think many are using it, and
4708 it is clearly obsolete.
4710 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
4713 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
4714 client certificate was expired. A simple patch fixes this, though I don't
4715 understand the full logic of why the verify callback is called multiple
4718 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
4719 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
4720 which causes problems with some clients (such as the Certicom SSL Plus
4721 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
4722 disables the coutermeasure allowing Eudora to connect."
4724 38. Exim was not checking that a write() to a log file succeeded. This could
4725 lead to Bad Things if a log got too big, in particular if it hit a file
4726 size limit. Exim now panics and dies if it cannot write to a log file, just
4727 as it does if it cannot open a log file.
4729 39. Modified OS/Makefile-Linux so that it now contains
4731 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
4733 The two -D definitions ensure that Exim is compiled with large file
4734 support, which makes it possible to handle log files that are bigger than
4737 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
4738 instance) a domain was checked against a named list that involved a lookup,
4739 causing $domain_data to be set, then another domain was checked against the
4740 same list, then the first domain was re-checked, the value of $domain_data
4741 after the final check could be wrong. In particular, if the second check
4742 failed, it could be set empty. This bug probably also applied to
4745 41. The strip_trailing_dot option was not being applied to the address given
4746 with the -f command-line option.
4748 42. The code for reading a message's header from the spool was incrementing
4749 $received_count, but never initializing it. This meant that the value was
4750 incorrect (doubled) while delivering a message in the same process in which
4751 it was received. In the most common configuration of Exim, this never
4752 happens - a fresh exec is done - but it can happen when
4753 deliver_drop_privilege is set.
4755 43. When Exim logs an SMTP synchronization error - client data sent too soon -
4756 it now includes up to 150 characters of the unexpected data in the log
4759 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
4760 and building data strings. The size of both of these buffers was 10 000
4761 bytes - far larger than anybody would *ever* want, thought I. Needless to
4762 say, somebody hit the limit. I have increased the maximum line length to
4763 20 000 and the maximum data length of concatenated lines to 100 000. I have
4764 also fixed two bugs, because there was no checking on these buffers. Tsk,
4765 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
4766 buffer is too small.
4768 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
4769 lsearch lookups. Now it does.
4771 46. When parsing a route_list item in a manualroute router, a fixed-length
4772 buffer was used for the list of hosts. I made this 1024 bytes long,
4773 thinking that nobody would ever have a list of hosts that long. Wrong.
4774 Somebody had a whole pile of complicated expansion conditions, and the
4775 string was silently truncated, leading to an expansion error. It turns out
4776 that it is easier to change to an unlimited length (owing to other changes
4777 that have happened since this code was originally written) than to build
4778 structure for giving a limitation error. The length of the item that
4779 expands into the list of hosts is now unlimited.
4781 47. The lsearch lookup could not handle data where the length of text line was
4782 more than 4095 characters. Such lines were truncated, leading to shortened
4783 data being returned. It should now handle lines of any length.
4785 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
4786 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
4789 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
4790 replacing strings such as BIN_DIRECTORY in a source file: the replacement
4791 no longer happens in comment lines. A list of replacements is now placed
4792 at the head of all of the source files, except those whose only change is
4793 to replace PERL_COMMAND in the very first #! line.
4795 50. Replaced the slow insertion sort in queue.c, for sorting the list of
4796 messages on the queue, with a bottom-up merge sort, using code contributed
4797 by Michael Haardt. This should make operations like -bp somewhat faster on
4798 large queues. It won't affect queue runners, except when queue_run_in_order
4801 51. Installed eximstats 1.31 in the distribution.
4803 52. Added support for SRV lookups to the dnslookup router.
4805 53. If an ACL referred to $message_body or $message_body_end, the value was not
4806 reset for any messages that followed in the same SMTP session.
4808 54. The store-handling optimization for building very long strings was not
4809 differentiating between the different store pools. I don't think this
4810 actually made any difference in practice, but I've tidied it.
4812 55. While running the routers to verify a sender address, $sender_address
4813 was still set to the sender address. This is wrong, because when routing to
4814 send a bounce to the sender, it would be empty. Therefore, I have changed
4815 it so that, while verifying a sender address, $sender_address is set to <>.
4816 (There is no change to what happens when verifying a recipient address.)
4818 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
4819 target A or AAAA records (if not already returned) without resetting the
4820 qualify_single or search_parents options of the DNS resolver. These are
4821 inappropriate in this case because the targets of MX and SRV records must
4822 be FQDNs. A broken DNS record could cause trouble if it happened to have a
4823 target that, when qualified, matched something in the local domain. These
4824 two options are now turned off when doing these lookups.
4826 57. It seems that at least some releases of Reiserfs (which does not have the
4827 concept of a fixed number of inodes) returns zero and not -1 for the
4828 number of available inodes. This interacted badly with check_spool_inodes,
4829 which assumed that -1 was the "no such thing" setting. What I have done is
4830 to check that the total number of inodes is greater than zero before doing
4831 the test of how many are available.
4833 58. When a "warn" ACL statement has a log_message modifier, the message is
4834 remembered, and not repeated. This is to avoid a lot of repetition when a
4835 message has many recipients that cause the same warning to be written.
4836 Howewer, Exim was preserving the list of already written lines for an
4837 entire SMTP session, which doesn't seem right. The memory is now reset if a
4838 new message is started.
4840 59. The "rewrite" debugging flag was not showing the result of rewriting in the
4841 debugging output unless log_rewrite was also set.
4843 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
4844 of (int)(handle) when we know that handle contains (void *)(-1).
4846 61. The Exim daemon panic-logs an error return when it closes the incoming
4847 connection. However "connection reset by peer" seems to be common, and
4848 isn't really an error worthy of noting specially, so that particular error
4851 62. When Exim is trying to find all the local interfaces, it used to panic and
4852 die if the ioctl to get the interface flags failed. However, it seems that
4853 on at least one OS (Solaris 9) it is possible to have an interface that is
4854 included in the list of interfaces, but for which you get a failure error
4855 for this call. This happens when the interface is not "plumbed" into a
4856 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
4857 failure of the "get flags" call assumes that the interface is down.
4859 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
4860 makes life easier for people who are doing arithmetic on fields extracted
4861 from dates, where you often get leading zeros that should not be
4862 interpreted as octal.
4864 64. Added qualify_domain to the redirect router, to override the global
4867 65. If a pathologically long header line contained very many addresses (the
4868 report of this problem mentioned 10 000) and each of them was rewritten,
4869 Exim could use up a very large amount of memory. (It kept on making new
4870 copies of the header line as it rewrote, and never released the old ones.)
4871 At the expense of a bit more processing, the header rewriting function has
4872 been changed so that it no longer eats memory in this way.
4874 66. The generation of the Received: header has been moved from the time that a
4875 message starts to be received, to the time that it finishes. The timestamp
4876 in the Received: header should now be very close to that of the <= log
4877 line. There are two side-effects of this change:
4879 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
4880 logged header lines no longer include the local Received: line, because
4881 it has not yet been created. The same applies to a copy of the message
4882 that is returned to a non-SMTP sender when a message is rejected.
4884 (b) When a filter file is tested using -bf, no additional Received: header
4885 is added to the test message. After some thought, I decided that this
4888 This change does not affect the value of $received_for. It is still set
4889 after address rewriting, but before local_scan() is called.
4891 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
4893 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
4894 gave an unhelpful panic error message, and a defer error. I have managed to
4895 change this behaviour so that it now rejects any supplied certificate,
4896 which seems right, as the list of acceptable certificates is empty.
4898 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
4899 gave an unhelpful defer error. I have not managed to make this reject any
4900 supplied certificates, but the error message it gives is "no certificate
4901 supplied", which is not helpful.
4903 70. exigrep's output now also includes lines that are not associated with any
4904 message, but which match the given pattern. Implemented by a patch from
4905 Martin Sluka, which also tidied up the Perl a bit.
4907 71. Recipient callout verification, like sender verification, was using <> in
4908 the MAIL FROM command. This isn't really the right thing, since the actual
4909 sender may affect whether the remote host accepts the recipient or not. I
4910 have changed it to use the actual sender in the callout; this means that
4911 the cache record is now keyed on a recipient/sender pair, not just the
4912 recipient address. There doesn't seem to be a real danger of callout loops,
4913 since a callout by the remote host to check the sender would use <>.
4914 [SEE ABOVE: changed after hitting problems.]
4916 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
4917 temporary errors. However, in the case of such a code being given after
4918 the end of a data transmission (i.e. after ".") Exim was failing to write
4919 a retry record for the message. (Yes, there was some broken host that was
4920 actually sending 8xx at this point.)
4922 73. An unknown lookup type in a host list could cause Exim to panic-die when
4923 the list was checked. (An example that provoked this was putting <; in the
4924 middle of a list instead of at the start.) If this happened during a DATA
4925 ACL check, a -D file could be left lying around. This kind of configuration
4926 error no longer causes Exim to die; instead it causes a defer errror. The
4927 incident is still logged to the main and panic logs.
4929 74. Buglet left over from Exim 3 conversion. The message "too many messages
4930 in one connection" was written to the rejectlog but not the mainlog, except
4931 when address rewriting (yes!) was being logged.
4933 75. Added write_rejectlog option.
4935 76. When a system filter was run not as root (that is, when system_filter_user
4936 was set), the values of the $n variables were not being returned to the
4937 main process; thus, they were not subsequently available in the $sn
4940 77. Added +return_path_on_delivery log selector.
4942 78. A connection timeout was being treated differently from recipients deferred
4943 when testing hosts_max_try with a message that was older than the host's
4944 retry timeout. (The host should not be counted, thus allowing all hosts to
4945 be tried at least once before bouncing.) This may have been the cause of an
4946 occasionally reported bug whereby a message would remain on the queue
4947 longer than the retry timeout, but would be bounced if a delivery was
4948 forced. I say "may" because I never totally pinned down the problem;
4949 setting up timeout/retry tests is difficult. See also the next item.
4951 79. The ultimate address timeout was not being applied to errors that involved
4952 a combination of host plus message (for example, a timeout on a MAIL
4953 command). When an address resolved to a number of possible hosts, and they
4954 were not all tried for each delivery (e.g. because of hosts_max_try), a
4955 message could remain on the queue longer than the retry timeout.
4957 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4960 81. Fixed an obscure SMTP outgoing bug which required at least the following
4961 conditions: (a) there was another message waiting for the same server;
4962 (b) the server returned 5xx to all RCPT commands in the first message so
4963 that the message was not completed; (c) the server dropped the connection
4964 or gave a negative response to the RSET that Exim sends to abort the
4965 transaction. The observed case was a dropped connection after DATA that had
4966 been sent in pipelining mode. That is, the server had advertised PIPELINING
4967 but was not implementing it correctly. The effect of the bug was incorrect
4968 behaviour, such as trying another host, and this could lead to a crash.
4974 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4975 and daemon.c were passed as pointers to ints; they should have been
4976 pointers to socklen_t variables (which are typically unsigned ints).
4978 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4981 3. Fixed a really odd bug that affected only the testing scheme; patching a
4982 certain fixed string in the binary changed the value of another string that
4983 happened to be identical to the end of the original first string.
4985 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4986 name", it returns that address as the IP address. On some operating
4987 systems (e.g. Solaris), it also passes back the IP address string as the
4988 "host name". However, on others (e.g. Linux), it passes back an empty
4989 string. Exim wasn't checking for this, and was changing the host name to an
4990 empty string, assuming it had been canonicized.
4992 5. Although rare, it is permitted to have more than one PTR record for a given
4993 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4994 all the names associated with an address, because they do in Solaris.
4995 However, it seems that they do not in Linux for data that comes from the
4996 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4997 I found this out when I moved to a new Linux workstation and tried to run
4998 the Exim test suite.
5000 To get round this problem I have changed the code so that it now does its
5001 own call to the DNS to look up PTR records when searching for a host name.
5002 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5003 addresses that are only in /etc/hosts are still found.
5005 This behaviour is, however, controlled by an option called host_lookup_
5006 order, which defaults to "bydns:byaddr". If people want to use the other
5007 order, or indeed, just use one or the other means of lookup, they can
5008 specify it in this variable.
5010 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5011 some operating systems, this comes back from gethostbyaddr() as an empty
5012 string, and this is what Exim used to test for. However, it seems that in
5013 other systems, "." is yielded. Exim now tests for this case too.
5015 7. The values of check_spool_space and check_log_space are now held internally
5016 as a number of kilobytes instead of an absolute number of bytes. If a
5017 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5018 kilobyte. This means that much larger values can be stored.
5020 8. Exim monitor: an attempt to get the action menu when not actually pointing
5021 at a message produces an empty menu entitled "No message selected". This
5022 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5023 no entries in it ("Shell widget menu has zero width and/or height"). So I
5024 have added a single, blank menu entry in this case.
5026 9. Added ${quote_local_part.
5028 10. MIME decoding is now applied to the contents of Subject: header lines when
5031 11. Now that a reference to $sender_host_address automatically causes a reverse
5032 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5033 host lookup before query-style lookups in lists that might use this
5034 variable. This has therefore been abolished, and the "net-" prefix is no
5035 longer necessary for query-style lookups.
5037 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5038 have been a typo for LFLAGS, so it has been changed.
5040 13. The install script calls Exim with "-C /dev/null" in order to find the
5041 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5042 to be output. Howeve, since Exim outputs its version number before the
5043 error, it didn't break the script. It just looked ugly. I fixed this by
5044 always allowing "-C /dev/null" if the caller is root.
5046 14. Ignore overlarge ACL variable number when reading spool file - insurance
5047 against a later release with more variables having written the file.
5049 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5050 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5051 this, as well as the form without the "IPv6" on the front.
5053 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5054 OS/Makefile-Darwin file.
5056 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5057 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5059 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5060 variable. However, some OS use ints for inodes. I've added cast to long int
5061 to get rid of the compiler warning.
5063 19. I had forgotten to lock out "/../" in configuration file names when
5064 ALT_CONFIG_PREFIX was set.
5066 20. Routers used for verification do not need to specify transports. However,
5067 if such a router generated a host list, and callout was configured, Exim
5068 crashed, because it could not find a port number from the (non-existent)
5069 transport. It now assumes port 25 in this circumstance.
5071 21. Added the -t option to exigrep.
5073 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5074 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5075 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5076 them is set, LOOKUP_LSEARCH is forced.
5078 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5080 24. Added sender and host information to the "rejected by local_scan()" log
5081 line; previously there was no indication of these.
5083 25. Added .include_if_exists.
5085 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5086 Linux. It turns out that not all file systems support this. Apparently some
5087 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5088 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5089 not supported on the file descriptor, is now ignored when Exim is trying to
5090 sync a directory. This applies only to Linux.
5092 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5094 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5095 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5098 29. Added support for maildirsize files from supplied patch (modified a bit).
5100 30. The use of :fail: followed by an empty string could lead Exim to respond to
5101 sender verification failures with (e.g.):
5103 550 Verification failed for <xxx>
5104 550 Sender verify failed
5106 where the first response line was missing the '-' that indicates it is not
5107 the final line of the response.
5109 31. The loop for finding the name of the user that called Exim had a hardwired
5110 limit of 10; it now uses the value of finduser_retries, which is used for
5111 all other user lookups.
5113 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5116 33. Exim was neglecting to zero errno before one call of strtol() when
5117 expanding a string and expecting an integer value. On some systems this
5118 resulted in spurious "integer overflow" errors. Also, it was casting the
5119 result into an int without checking.
5121 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5122 did not work. The code looks as if it has *never* worked, though it appears
5123 to have been documented since at least releast 1.62. I have made it work.
5125 35. The "timeout_DNS" error in retry rules, also documented since at least
5126 1.62, also never worked. As it isn't clear exactly what this means, and
5127 clearly it isn't a major issue, I have abolished the feature by treating it
5128 as "timeout", and writing a warning to the main and panic logs.
5130 36. The display of retry rules for -brt wasn't always showing the error code
5133 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5134 timeout_connect_A, timeout_connect_MX.
5136 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5137 to the empty sender.
5139 39. The daemon was not analysing the content of -oX till after it had closed
5140 stderr and disconnected from the controlling terminal. This meant that any
5141 syntax errors were only noted on the panic log, and the return code from
5142 the command was 0. By re-arranging the code a little, I've made the
5143 decoding happen first, so such errors now appear on stderr, and the return
5144 code is 1. However, the actual setting up of the sockets still happens in
5145 the disconnected process, so errors there are still only recorded on the
5148 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5149 connections (as happens on some IP stacks) was logged at start up time as
5150 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5151 it from "IPv6 and IPv4", which means that two separate sockets are being
5154 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5155 says whether AF_INET or AF_INET6 was passed as an argument.
5157 42. Exiwhat output was messed up when time zones were included in log
5160 43. Exiwhat now gives more information about the daemon's listening ports,
5161 and whether -tls-on-connect was used.
5163 44. The "port" option of the smtp transport is now expanded.
5165 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5166 silently ignored. Now an error message is written to the main and panic
5169 46. There's a new ACL modifier called "logwrite" which writes to a log file
5170 as soon as it is encountered.
5172 47. Added $local_user_uid and $local_user_gid at routing time.
5174 48. Exim crashed when trying to verify a sender address that was being
5177 49. Exim was recognizing only a space character after ".include". It now also
5178 recognizes a tab character.
5180 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5181 extracting the relevant information from the specification. The man page no
5182 longer contains scrambled data for the -d option, and I've added a section
5183 at the front about calling Exim under different names.
5185 51. Added "extra_headers" argument to the "mail" command in filter files.
5187 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5190 53. Installed eximstats 1.29.
5192 54. Added transport_filter_timeout as a generic transport option.
5194 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5195 Cc: header lines. This was required by RFC 822, but it not required by RFC
5198 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5199 incoming messages that did not have them. Now it does so only if the
5200 message originates locally, that is, if there is no associated remote host
5201 address. When Resent- header lines are present, this applies to the Resent-
5202 lines rather than the non-Resent- lines.
5204 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5205 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5207 58. Messages for configuration errors now include the name of the main
5208 configuration file - useful now that there may be more than one file in a
5209 list (.included file names were always shown).
5211 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5212 for those rare installations that do not start the daemon as root or run it
5213 setuid root. I've cut out the call to initgroups() if the daemon is not
5216 60. The Exim user and group can now be bound into the binary as text strings
5217 that are looked up at the start of Exim's processing.
5219 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5221 62. Added $mailstore_basename variable.
5223 63. Installed patch to sieve.c from Michael Haardt.
5225 64. When Exim failed to open the panic log after failing to open the main log,
5226 the original message it was trying to log was written to stderr and debug
5227 output, but if they were not available (the usual case in production), it
5228 was lost. Now it is written to syslog before the two lines that record the
5229 failures to open the logs.
5231 65. Users' Exim filters run in subprocesses under the user's uid. It is
5232 possible for a "deliver" command or an alias in a "personal" command to
5233 provoke an address rewrite. If logging of address rewriting is configured,
5234 this fails because the process is not running as root or exim. There may be
5235 a better way of dealing with this, but for the moment (because 4.30 needs
5236 to be released), I have disabled address rewrite logging when running a
5237 filter in a non-root, non-exim process.
5243 1. The buildconfig auxiliary program wasn't quoting the value set for
5244 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5245 not defined. This bug was masked in 4.22 by the effect that was fixed in
5248 2. Some messages that were rejected after a message id was allocated were
5249 shown as "incomplete" by exigrep. It no longer does this for messages that
5250 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5252 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5253 have allow_domain_literals set, the ID did not get logged in the <= line.
5254 Domain literals are now always recognized in Message-ID: header lines.
5256 4. The first argument for a ${extract expansion item is the key name or field
5257 number. Leading and trailing spaces in this item were not being ignored,
5258 causing some misleading effects.
5260 5. When deliver_drop_privilege was set, single queue runner processes started
5261 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5262 same command in the process it spins off) were not dropping privilege.
5264 6. When the daemon running as "exim" started a queue runner, it always
5265 re-executed Exim in the spun-off process. This is a waste of effort when
5266 deliver_drop_privilege is set. The new process now just calls the
5267 queue-runner function directly.
5273 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5276 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5277 ignored. Though the use of -oP was forcing the writing of a pid file, it
5278 was always written to the default place.
5280 3. If the message "no IP address found for host xxxx" is generated during
5281 incoming verification, it is now followed by identification of the incoming
5282 connection (so you can more easily find what provoked it).
5284 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5286 5. Added some features to "harden" Exim a bit more against certain attacks:
5288 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5289 be put in Local/Makefile. This is like the never_users runtime option,
5290 but it cannot be overridden. The default setting is "root".
5292 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5293 prefix string with which any file named in a -C command line option
5296 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5297 is retained for -C and -D only if the caller of Exim is root. Without
5298 it, the exim user may also use -C and -D and retain privilege.
5300 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5301 command line option is disabled.
5303 6. Macro names set by the -D option must start with an upper case letter, just
5304 like macro names defined in the configuration file.
5306 7. Added "dereference=" facility to LDAP.
5308 8. Two instances of the typo "uknown" in the source files are fixed.
5310 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5311 the Configure-Makefile script screwed up while processing it.
5313 10. Incorporated PCRE 4.4.
5315 11. The SMTP synchronization check was not operating right at the start of an
5316 SMTP session. For example, it could not catch a HELO sent before the client
5317 waited for the greeting. There is now a check for outstanding input at the
5318 point when the greeting is written. Because of the duplex, asynchronous
5319 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5320 way, but not yet received, when the check is performed.
5322 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5323 on TCP/IP sockets, because this apparently causes some broken clients to
5326 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5327 unchanged) from the Cygwin maintainer.
5329 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5330 was supported instead of testing for mbx. Effectively a typo.
5332 15. The spa authenticator server code was not checking that the input it
5333 received was valid base64.
5335 16. The debug output line for the "set" modifier in ACLs was not showing the
5336 name of the variable that was being set.
5338 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5340 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5341 lookup on its own. Something else was needed to trigger the lookup. For
5342 example, a match in host_lookup or the need for a host name in a host list.
5343 Now, if $sender_host_name is referenced and the host name has not yet been
5344 looked up, a lookup is performed. If the lookup fails, the variable remains
5345 empty, and $host_lookup_failed is set to "1".
5347 19. Added "eqi" as a case-independent comparison operator.
5349 20. The saslauthd authentication condition could segfault if neither service
5350 nor realm was specified.
5352 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5353 error message that was logged was misleading, and incoming SMTP
5354 connections were dropped. The message is now more accurate, and temporary
5355 errors are given to SMTP connections.
5357 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5358 (except RSET) with a 421 error, until QUIT is received. However, it was
5359 failing to send a response to QUIT.
5361 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5362 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5363 ACL is now run after the helo_try_verify_hosts code.
5365 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5366 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5367 case-independent, so other case variants are also recognized). Apparently
5368 some systems use these upper case variants.
5370 25. If more than two messages were waiting for the same host, and a transport
5371 filter was specified for the transport, Exim sent two messages over the
5372 same TCP/IP connection, and then failed with "socket operation on non-
5373 socket" when it tried to send the third.
5375 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5377 27. The extern definition of crypt16() in expand.c was not being excluded when
5378 the OS had its own crypt16() function.
5380 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5381 as a preferred synonym for return_size_limit, both as an option and as an
5384 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5386 30. Changed the default configuration ACL to relax the local part checking rule
5387 for addresses that are not in any local domains. For these addresses,
5388 slashes and pipe symbols are allowed within local parts, but the sequence
5389 /../ is explicitly forbidden.
5391 31. SPA server authentication was not clearing the challenge buffer before
5394 32. log_message in a "warn" ACL statement was writing to the reject log as
5395 well as to the main log, which contradicts the documentation and doesn't
5396 seem right (because no rejection is happening). So I have stopped it.
5398 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5399 However, I am unable to do any testing of this.
5401 34. Fixed an infelicity in the appendfile transport. When checking directories
5402 for a mailbox, to see if any needed to be created, it was accidentally
5403 using path names with one or more superfluous leading slashes; tracing
5404 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5406 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5407 contents are added to the log line that is written for every discarded
5408 recipient. (Previously a log_message setting was ignored.)
5410 36. The ${quote: operator now quotes the string if it is empty.
5412 37. The install script runs exim in order to find its version number. If for
5413 some reason other than non-existence or emptiness, which it checks, it
5414 could not run './exim', it was installing it with an empty version number,
5415 i.e. as "exim-". This error state is now caught, and the installation is
5418 38. An argument was missing from the function that creates an error message
5419 when Exim fails to connect to the socket for saslauthd authentication.
5420 This could cause Exim to crash, or give a corrupted message.
5422 39. Added isip, isip4, and isip6 to ${if conditions.
5424 40. The ACL variables $acl_xx are now saved with the message, and can be
5425 accessed later in routers, transports, and filters.
5427 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
5428 strings in the file are not string-expanded.
5430 42. If a MAIL command specified a SIZE value that was too large to fit into an
5431 int variable, the check against message_size_limit failed. Such values are
5432 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
5433 day this will have to be increased, but I don't think I want to be around
5434 when emails are that large.
5441 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
5442 iconv() is not standard in FreeBSD.
5444 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
5445 IPv6 enabled. The observed symptom was a segmentation fault on return from
5446 the function os_common_find_running_interfaces() in src/os.c.
5448 3. In the check_special_case() function in daemon.c I had used "errno" as an
5449 argument name, which causes warnings on some systems. This was basically a
5450 typo, since it was named "eno" in the comments!
5452 4. The code that waits for the clock to tick (at a resolution of some fraction
5453 of a second) so as to ensure message-id uniqueness was always waiting for
5454 at least one whole tick, when it could have waited for less. [This is
5455 almost certainly not relevant at current processor speeds, where it is
5456 unlikely to ever wait at all. But we try to future-proof.]
5458 5. The function that sleeps for a time interval that includes fractions of a
5459 second contained a race. It did not block SIGALRM between setting the
5460 timer, and suspending (a couple of lines later). If the interval was short
5461 and the sigsuspend() was delayed until after it had expired, the suspension
5462 never ended. On busy systems this could lead to processes getting stuck for
5465 6. Some uncommon configurations may cause a lookup to happen in a queue runner
5466 process, before it forks any delivery processes. The open lookup caching
5467 mechanism meant that the open file or database connection was passed into
5468 the delivery process. The problem was that delivery processes always tidy
5469 up cached lookup data. This could cause a problem for the next delivery
5470 process started by the queue runner, because the external queue runner
5471 process does not know about the closure. So the next delivery process
5472 still has data in the lookup cache. In the case of a file lookup, there was
5473 no problem because closing a file descriptor in a subprocess doesn't affect
5474 the parent. However, if the lookup was caching a connection to a database,
5475 the connection was closed, and the second delivery process was likely to
5476 see errors such as "PGSQL: query failed: server closed the connection
5477 unexpectedly". The problem has been fixed by closing all cached lookups
5478 in a queue runner before running a delivery process.
5480 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
5481 seem to have the "const" qualifier which it has on other OS. I've
5484 8. Change 4.21/2 was too strict. It is only if there are two authenticators
5485 *of the same type* (client or server) with the same public name that an
5486 error should be diagnosed.
5488 9. When Exim looked up a host name for an IP address, but failed to find the
5489 original IP address when looking up the host name (a safety check), it
5490 output the message "<ip address> does not match any IP for NULL", which was
5491 confusing, to say the least. The bug was that the host name should have
5492 appeared instead of "NULL".
5494 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
5495 user that is built into the binary, and the -C or -D options is used, root
5496 privilege is dropped before the configuration file is read. In addition,
5497 logging is switched to stderr instead of the normal log files. If the
5498 configuration then re-defines the Exim user, the unprivileged environment
5499 is probably not what is expected, so Exim logs a panic warning message (but
5502 However, if deliver_drop_privilege is set, the unprivileged state may well
5503 be exactly what is intended, so the warning has been cut out in that case,
5504 and Exim is allowed to try to write to its normal log files.
5510 1. smtp_return_error_details was not giving details for temporary sender
5511 or receiver verification errors.
5513 2. Diagnose a configuration error if two authenticators have the same public
5516 3. Exim used not to create the message log file for a message until the first
5517 delivery attempt. This could be confusing when incoming messages were held
5518 for policy or load reasons. The message log file is now created at the time
5519 the message is received, and an initial "Received" line is written to it.
5521 4. The automatically generated man page for command line options had a minor
5522 bug that caused no ill effects; however, a more serious problem was that
5523 the procedure for building the man page automatically didn't always
5524 operate. Consequently, release 4.20 contains an out-of-date version. This
5525 shouldn't happen again.
5527 5. When building Exim with embedded Perl support, the script that builds the
5528 Makefile was calling 'perl' to find its compile-time parameters, ignoring
5529 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
5531 6. The freeze_tell option was not being used for messages that were frozen on
5532 arrival, either by an ACL or by local_scan().
5534 7. Added the smtp_incomplete_transaction log selector.
5536 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
5537 was accepting AUTH without a new EHLO.
5539 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
5540 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
5543 10. Logging of TCP/IP connections (when configured) now happens in the main
5544 daemon process instead of the child process, so that the TCP/IP connection
5545 count is more accurate (but it can never be perfect).
5547 11. The use of "drop" in a nested ACL was not being handled correctly in the
5548 outer ACL. Now, if condition failure induced by the nested "drop" causes
5549 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
5550 or "require"), the connection is dropped.
5552 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
5553 that yield "discard" can now be used with an "accept" or a "discard" verb,
5554 but an error is generated for any others (because I can't see a useful way
5555 to define what should happen).
5557 13. When an ACL is read dynamically from a file (or anywhere else), the lines
5558 are now processed in the same way as lines in the Exim configuration file.
5559 In particular, continuation lines are supported.
5561 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
5563 15. Added -ti meaning -t -i.
5565 16. Check for letters, digits, hyphens, and dots in the names of dnslist
5566 domains, and warn by logging if others are found.
5568 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
5569 returned from GIFCONF when Exim is trying to find the list of interfaces on
5570 a host. The code in os.c has been modified to copy each ifreq to an aligned
5571 structure in all cases.
5573 Also, in some cases, the returned ifreq's were being copied to a 'struct
5574 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
5575 means the last couple of bytes of an IPv6 address could be chopped if the
5576 ifreq contained only a normal sockaddr (14 bytes storage).
5578 18. Named domain lists were not supported in the hosts_treat_as_local option.
5579 An entry such as +xxxx was not recognized, and was treated as a literal
5582 19. Ensure that header lines added by a DATA ACL are included in the reject log
5583 if the ACL subsequently rejects the message.
5585 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
5586 MD5 (which is deprecated).
5588 21. When testing a filter file using -bf, Exim was writing a message when it
5589 took the sender from a "From " line in the message, but it was not doing so
5590 when it took $return_path from a Return-Path: header line. It now does.
5592 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
5593 with a valid header line field name (a series of printing characters
5594 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
5596 23. Changed "disc" in the source to "disk" to conform to the documentation and
5597 the book and for uniformity.
5599 24. Ignore Sendmail's -Ooption=value command line item.
5601 25. When execve() failed while trying to run a command in a pipe transport,
5602 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
5603 could be confused with a return value of 69 from the command itself. This
5604 has been changed to 127, the value the shell returns if it is asked to run
5605 a non-existent command. The wording for the related log line suggests a
5606 non-existent command as the problem.
5608 26. If received_header_text expands to an empty string, do not add a Received:
5609 header line to the message. (Well, it adds a token one on the spool, but
5610 marks it "old" so that it doesn't get used or transmitted.)
5612 27. Installed eximstats 1.28 (addition of -nt option).
5614 28. There was no check for failure on the call to getsockname() in the daemon
5615 code. This can fail if there is a shortage of resources on the system, with
5616 ENOMEM, for example. A temporary error is now given on failure.
5618 29. Contrary to the C standard, it seems that in some environments, the
5619 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
5620 program. Exim now does this explicitly; it affects the formatting of
5621 timestamps using strftime().
5623 30. If exiqsumm was given junk data, it threw up some uninitialized variable
5624 complaints. I've now initialized all the variables, to avoid this.
5626 32. Header lines added by a system filter were not being "seen" during
5627 transport-time rewrites.
5629 33. The info_callback() function passed to OpenSSL is set up with type void
5630 (*)(SSL *, int, int), as described somewhere. However, when calling the
5631 function (actually a macro) that sets it up, the type void(*)() is
5632 expected. I've put in a cast to prevent warnings from picky compilers.
5634 34. If a DNS black list lookup found a CNAME record, but there were no A
5635 records associated with the domain it pointed at, Exim crashed.
5637 35. If a DNS black list lookup returned more than one A record, Exim ignored
5638 all but the first. It now scans all returned addresses if a particular IP
5639 value is being sought. In this situation, the contents of the
5640 $dnslist_value variable are a list of all the addresses, separated by a
5643 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
5644 to accept a host name and all its aliases if the forward lookup for any of
5645 them yielded the IP address of the incoming connection. Now it accepts only
5646 those names whose forward lookup yields the correct IP address. Any other
5647 names are discarded. This closes a loophole whereby a rogue DNS
5648 administrator could create reverse DNS records to break through a
5649 wildcarded host restriction in an ACL.
5651 37. If a user filter or a system filter that ran in a subprocess used any of
5652 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
5653 the wrong values were passed to the pipe command ($thisaddress had the
5654 value of $0, $0 had the value of $1, etc). This bug was introduced by
5655 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
5657 38. Improved the line breaking for long SMTP error messages from ACLs.
5658 Previously, if there was no break point between 40 and 75 characters, Exim
5659 left the rest of the message alone. Two changes have been made: (a) I've
5660 reduced the minimum length to 35 characters; (b) if it can't find a break
5661 point between 35 and 75 characters, it looks ahead and uses the first one
5662 that it finds. This may give the occasional overlong line, but at least the
5663 remaining text gets split now.
5665 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
5666 file descriptors might be low, and that setting 1000 would always raise it.
5667 It turns out that in some environments, the limit is already over 1000 and
5668 that lowering it causes trouble. So now Exim takes care not to decrease it.
5670 40. When delivering a message, the value of $return_path is set to $sender_
5671 address at the start of routing (routers may change the value). By an
5672 oversight, this default was not being set up when an address was tested by
5673 -bt or -bv, which affected the outcome if any router or filter referred to
5676 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
5677 log only when "message" or "log_message" are set. However, if one of the
5678 conditions was an address verification, or a call to a nested ACL, the
5679 messages generated by the underlying test were being passed through. This
5680 no longer happens. The underlying message is available in $acl_verify_
5681 message for both "message" and "log_message" expansions, so it can be
5682 passed through if needed.
5684 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
5685 new expansion $bh_ to give the encoded byte string without charset
5686 translation. Translation happens only if iconv() is available; HAVE_ICONV
5687 indicates this at build time. HEADERS_CHARSET gives the charset to
5688 translate to; headers_charset can change it in the configuration, and
5689 "headers charset" can change it in an individual filter file.
5691 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
5692 that creates RFC 2047 encoded "words" labels them as that charset instead
5693 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
5694 expansion operator; (ii) when Exim creates a From: line for a local
5695 message; (iii) when a header line is rewritten to include a "phrase" part.
5697 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
5698 buggy, causing it to skip the first lines of messages whose message ID
5699 ended in 'D'. This would not have bitten before Exim release 4.14, because
5700 message IDs were unlikely to end in 'D' before then. The effect was to have
5701 incorrect size information for certain domains.
5703 45. #include "config.h" was missing at the start of the crypt16.c module. This
5704 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
5707 46. If there was a timeout during a "random" callout check, Exim treated it as
5708 a failure of the random address, and carried on sending RSET and the real
5709 address. If the delay was just some slowness somewhere, the response to the
5710 original RCPT would be taken as a response to RSET and so on, causing
5711 mayhem of various kinds.
5713 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
5714 when I implemented it. It didn't allow for the fact that some option values
5715 may legitimatetly be negative (e.g. size_addition), and it didn't even do
5716 the right test for positive values.
5718 48. Domain names in DNS records are case-independent. Exim always looks them up
5719 in lower case. Some resolvers return domain names in exactly the case they
5720 appear in the zone file, that is, they may contain uppercase letters. Not
5721 all resolvers do this - some return always lower case. Exim was treating a
5722 change of case by a resolver as a change of domain, similar to a widening
5723 of a domain abbreviation. This triggered its re-routing code and so it was
5724 trying to route what was effectively the same domain again. This normally
5725 caused routing to fail (because the router wouldn't handle the domain
5726 twice). Now Exim checks for this case specially, and just changes the
5727 casing of the domain that it ultimately uses when it transmits the message
5730 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
5733 50. If a filter generated a file delivery with a non-absolute name (possible if
5734 no home directory exists for the router), the forbid_file option was not
5737 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
5738 the existing equality matching.
5740 52. Exim was using ints instead of ino_t variables in some places where it was
5741 dealing with inode numbers.
5743 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
5744 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
5745 called TMPDIR, and if it finds it is different, it changes its value.
5747 54. The smtp_printf() function is now made available to local_scan() so
5748 additional output lines can be written before returning. There is also an
5749 smtp_fflush() function to enable the detection of a dropped connection.
5750 The variables smtp_input and smtp_batched_input are exported to
5753 55. Changed the default runtime configuration: the message "Unknown user"
5754 has been removed from the ACL, and instead placed on the localuser router,
5755 using the cannot_route_message feature. This means that any verification
5756 failures that generate their own messages won't get overridden. Similarly,
5757 the "Unrouteable address" message that was in the ACL for unverifiable
5758 relay addresses has also been removed.
5760 56. Added hosts_avoid_esmtp to the smtp transport.
5762 57. The exicyclog script was not checking for the esoteric option
5763 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
5764 will work only if exicyclog is run under the appropriate euid.
5766 58. Following a discussion on the list, the rules by which Exim recognises line
5767 endings on incoming messages have been changed. The -dropcr and drop_cr
5768 options are now no-ops, retained only for backwards compatibility. The
5769 following line terminators are recognized: LF CRLF CR. However, special
5770 processing applies to CR:
5772 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
5773 nor a local message in the state where . is a terminator.
5775 (ii) If a bare CR is encountered in a header line, an extra space is added
5776 after the line terminator so as not to end the header. The reasoning
5777 behind this is that bare CRs in header lines are most likely either
5778 to be mistakes, or people trying to play silly games.
5780 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
5781 was being incorrectly given as 18 bytes larger than it should have been.
5782 This is a VOB (very old bug).
5784 60. This may never have affected anything current, but just in case it has:
5785 When the local host is found other than at the start of a list of hosts,
5786 the local host, those with the same MX, and any that follow, are discarded.
5787 When the list in question was part of a longer list of hosts, the following
5788 hosts (not currently being processed) were also being discarded. This no
5789 longer happens. I'm not sure if this situation could ever has previously
5792 61. Added the "/MX" feature to lists of hosts in the manualroute and query
5795 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
5796 header. This is something that is recommended in a new Internet Draft, and
5797 is something that is documented as being done by Sendmail. There are two
5798 possible values. For messages generated by the autoreply transport, Exim
5801 Auto-Submitted: auto-replied
5803 whereas for all other generated messages (e.g. bounces) it adds
5805 Auto-Submitted: auto-generated
5807 63. The "personal" condition in filters now includes a test for the
5808 Auto-Submitted: header. If it contains the string "auto-" the message it
5809 not considered personal.
5811 64. Added rcpt_include_affixes as a generic transport option.
5813 65. Added queue_only_override (default true).
5815 66. Added the syslog_duplication option.
5817 67. If what should have been the first header line of a message consisted of
5818 a space followed by a colon, Exim was mis-interpreting it as a header line.
5819 It isn't of course - it is syntactically invalid and should therefore be
5820 treated as the start of the message body. The misbehaviour could have
5821 caused a number of strange effects, including loss of data in subsequent
5822 header lines, and spool format errors.
5824 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
5825 client host had authenticated. This control can now be exercised by an ACL
5826 for more flexibility.
5828 69. By default, callouts do not happen when testing with -bh. There is now a
5829 variant, -bhc, which does actually run the callout code, including
5830 consulting and updating the callout cache.
5832 70. Added support for saslauthd authentication, courtesy of Alexander
5835 71. If statvfs() failed on the spool or log directories while checking their
5836 size for availability, Exim confusingly gave the error "space shortage".
5837 Furthermore, in debugging mode it crashed with a floating point exception.
5838 These checks are done if check_{spool,log}_{space,inodes} are set, and when
5839 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
5840 serious problem, Exim now writes to the main and panic logs when this
5841 happens, with details of the failure. It then refuses to accept the
5842 incoming message, giving the message "spool directory problem" or "log
5843 directory problem" with a 421 code for SMTP messages.
5845 72. When Exim is about to re-exec itself, it ensures that the file descriptors
5846 0, 1, and 2 exist, because some OS complain for execs without them (see
5847 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
5848 descriptors. However, the code omitted to check that the open succeeded,
5849 causing mysterious errors if for some reason the permissions on /dev/null
5850 got screwed. Now Exim writes a message to the main and panic logs, and
5851 bombs out if it can't open /dev/null.
5853 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
5854 interact so that it is all more flexible. It is supposed to remain
5855 backwards compatible. Also added extra_local_interfaces.
5857 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
5858 to bomb out with an assertion failure - to the client this appears as a
5859 connection drop. This problem occurs in the part of the code that was taken
5860 from the Samba project. Fortunately, the assertion is in a very simple
5861 function, so I have fixed this by reproducing the function inline in the
5862 one place where it is called, and arranging for authentication to fail
5863 instead of killing the process with assert().
5865 75. The SPA client code was not working when the server requested OEM rather
5866 than Unicode encoding.
5868 76. Added code to make require_files with a specific uid setting more usable in
5869 the case where statting the file as root fails - usually a non-root-mounted
5870 NFS file system. When this happens and the failure is EACCES, Exim now
5871 forks a subprocess and does the per-uid checking as the relevant uid.
5873 77. Added process_log_path.
5875 78. If log_file_path was not explicitly set, a setting of check_log_space or
5876 check_log_inodes was ignored.
5878 79. If a space check for the spool or log partitions fails, the incident is now
5879 logged. Of course, in the latter case the data may get lost...
5881 80. Added the %p formatting code to string_format() so that it can be used to
5882 print addresses in debug_print(). Adjusted all the address printing in the
5883 debugging in store.c to use %p rather than %d.
5885 81. There was a concern that a line of code in smtp_in.c could overflow a
5886 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
5887 initially expressed, the concern was not well-founded, because trailing
5888 spaces are removed early. However, if the trailing spaces were followed by
5889 a NULL, they did not get removed, so the overflow was possible. Two fixes
5892 (a) I re-wrote the offending code in a cleaner fashion.
5893 (b) If an incoming SMTP command contains a NULL character, it is rejected
5896 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
5897 runs initgroups(), so that if the Exim user is in any additional groups,
5898 they will be used during message reception.
5904 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / heiko / exim.git / blob