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.
249 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
250 hostname or reverse DNS when processing a host list. Used suggestions
251 from multiple comments on this bug.
253 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
259 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
261 This, or similar/improved, will also be change PP/11 of 4.82.
267 PP/01 Handle short writes when writing local log-files.
268 In practice, only affects FreeBSD (8 onwards).
269 Bugzilla 1053, with thanks to Dmitry Isaikin.
271 NM/01 Bugzilla 949 - Documentation tweak
273 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
276 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
278 PP/02 Implemented gsasl authenticator.
280 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
282 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
283 `pkg-config foo` for cflags/libs.
285 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
286 with rest of GSASL and with heimdal_gssapi.
288 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
289 `pkg-config foo` for cflags/libs for the TLS implementation.
291 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
292 properties get this fed in as external SSF. A number of robustness
293 and debugging improvements to the cyrus_sasl authenticator.
295 PP/08 cyrus_sasl server now expands the server_realm option.
297 PP/09 Bugzilla 1214 - Log authentication information in reject log.
298 Patch by Jeremy Harris.
300 PP/10 Added dbmjz lookup type.
302 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
304 PP/12 MAIL args handles TAB as well as SP, for better interop with
305 non-compliant senders.
306 Analysis and variant patch by Todd Lyons.
308 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
309 Bug report from Lars Müller <lars@samba.org> (via SUSE),
310 Patch from Dirk Mueller <dmueller@suse.com>
312 PP/13 tls_peerdn now print-escaped for spool files.
313 Observed some $tls_peerdn in wild which contained \n, which resulted
314 in spool file corruption.
316 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
317 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
318 or write after TLS renegotiation, which otherwise led to messages
321 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
322 as a tracking header (ie: a signed header comes before the signature).
323 Patch from Wolfgang Breyha.
325 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
326 comma-sep list; embedded commas doubled.
328 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
330 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
332 Report and patch from Dmitry Banschikov.
334 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
335 Removed SSL_clear() after SSL_new() which led to protocol negotiation
336 failures. We appear to now support TLS1.1+ with Exim.
338 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
339 lets Exim select keys and certificates based upon TLS SNI from client.
340 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
341 before an outbound SMTP session. New log_selector, +tls_sni.
343 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
344 NULL dereference. Report and patch from Alun Jones.
346 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
347 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
348 Not seeing resolver debug output on NetBSD, but suspect this is a
349 resolver implementation change.
351 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
352 Left warnings. Added "eximon gdb" invocation mode.
354 PP/21 Defaulting "accept_8bitmime" to true, not false.
356 PP/22 Added -bw for inetd wait mode support.
358 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
359 locate the relevant includes and libraries. Made this the default.
361 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
362 Bugzilla 1246, report and most of solution from Tomasz Kusy.
364 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
365 This may cause build issues on older platforms.
367 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
368 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
369 gnutls_require_mac & gnutls_require_protocols (no longer supported).
370 Added SNI support via GnuTLS too.
371 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
373 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
375 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
376 Applied second patch from Janne, implementing suggestion to default
377 multiple-strings-in-record handling to match SPF spec.
379 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
381 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
382 read-only, out of scope).
383 Patch from Wolfgang Breyha, report from Stuart Northfield.
385 PP/29 Fix three issues highlighted by clang analyser static analysis.
386 Only crash-plausible issue would require the Cambridge-specific
387 iplookup router and a misconfiguration.
388 Report from Marcin Mirosław.
390 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
392 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
393 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
394 As part of this, removing so much warning spew let me fix some minor
395 real issues in debug logging.
397 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
398 assignment on my part. Fixed.
400 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
401 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
402 Janne Snabb (who went above and beyond: thank you).
404 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
405 string otherwise requires a connection and a bunch more work and it's
406 relatively easy to get wrong. Should also expose TLS library linkage
409 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
410 64-bit ${eval} (JH/03).
412 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
413 GNU libc to support some of the 64-bit stuff, should not lead to
414 conflicts. Defined before os.h is pulled in, so if a given platform
415 needs to override this, it can.
417 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
418 protection layer was required, which is not implemented.
419 Bugzilla 1254, patch from Wolfgang Breyha.
421 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
422 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
423 tls_dhparam take prime identifiers. Also unbreak combination of
424 OpenSSL+DH_params+TLSSNI.
426 PP/39 Disable SSLv2 by default in OpenSSL support.
432 PP/01 Solaris build fix for Oracle's LDAP libraries.
433 Bugzilla 1109, patch from Stephen Usher.
435 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
437 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
440 TF/02 Fix a couple more cases where we did not log the error message
441 when unlink() failed. See also change 4.74-TF/03.
443 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
444 lock up or crash if it happened to be inside a call to libc when it
445 got a SIGUSR1 from exiwhat.
447 The SIGUSR1 handler appends the current process status to the process
448 log which is later printed by exiwhat. It used to use the general
449 purpose logging code to do this, but several functions it calls are
450 not safe for signals.
452 The new output code in the SIGUSR1 handler is specific to the process
453 log, and simple enough that it's easy to inspect for signal safety.
454 Removing some special cases also simplifies the general logging code.
455 Removing the spurious timestamps from the process log simplifies
458 TF/04 Improved ratelimit ACL condition.
460 The /noupdate option has been deprecated in favour of /readonly which
461 has clearer semantics. The /leaky, /strict, and /readonly update modes
462 are mutually exclusive. The update mode is no longer included in the
463 database key; it just determines when the database is updated. (This
464 means that when you upgrde Exim will forget old rate measurements.)
466 Exim now checks that the per_* options are used with an update mode that
467 makes sense for the current ACL. For example, when Exim is processing a
468 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
469 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
470 must specify per_mail/readonly. If you omit the update mode it defaults to
471 /leaky where that makes sense (as before) or /readonly where required.
473 The /noupdate option is now undocumented but still supported for
474 backwards compatibility. It is equivalent to /readonly except that in
475 ACLs where /readonly is required you may specify /leaky/noupdate or
476 /strict/noupdate which are treated the same as /readonly.
478 A useful new feature is the /count= option. This is a generalization
479 of the per_byte option, so that you can measure the throughput of other
480 aggregate values. For example, the per_byte option is now equivalent
481 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
483 The per_rcpt option has been generalized using the /count= mechanism
484 (though it's more complicated than the per_byte equivalence). When it is
485 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
486 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
487 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
488 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
489 Note that using per_rcpt with a non-readonly update mode in more than
490 one ACL will cause the recipients to be double-counted. (The per_mail
491 and per_byte options don't have this problem.)
493 The handling of very low rates has changed slightly. If the computed rate
494 is less than the event's count (usually one) then this event is the first
495 after a long gap. In this case the rate is set to the same as this event's
496 count, so that the first message of a spam run is counted properly.
498 The major new feature is a mechanism for counting the rate of unique
499 events. The new per_addr option counts the number of different
500 recipients that someone has sent messages to in the last time period. It
501 behaves like per_rcpt if all the recipient addresses are different, but
502 duplicate recipient addresses do not increase the measured rate. Like
503 the /count= option this is a general mechanism, so the per_addr option
504 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
505 example, measure the rate that a client uses different sender addresses
506 with the options per_mail/unique=$sender_address. There are further
507 details in the main documentation.
509 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
511 TF/06 Removed a few PCRE remnants.
513 TF/07 Automatically extract Exim's version number from tags in the git
514 repository when doing development or release builds.
516 PP/02 Raise smtp_cmd_buffer_size to 16kB.
517 Bugzilla 879. Patch from Paul Fisher.
519 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
520 Heavily based on revision 40f9a89a from Simon Arlott's tree.
523 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
525 PP/05 Variable $av_failed, true if the AV scanner deferred.
526 Bugzilla 1078. Patch from John Horne.
528 PP/06 Stop make process more reliably on build failure.
529 Bugzilla 1087. Patch from Heiko Schlittermann.
531 PP/07 Make maildir_use_size_file an _expandable_ boolean.
532 Bugzilla 1089. Patch from Heiko Schlittermann.
534 PP/08 Handle ${run} returning more data than OS pipe buffer size.
535 Bugzilla 1131. Patch from Holger Weiß.
537 PP/09 Handle IPv6 addresses with SPF.
538 Bugzilla 860. Patch from Wolfgang Breyha.
540 PP/10 GnuTLS: support TLS 1.2 & 1.1.
542 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
545 PP/11 match_* no longer expand right-hand-side by default.
546 New compile-time build option, EXPAND_LISTMATCH_RHS.
547 New expansion conditions, "inlist", "inlisti".
549 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
551 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
553 PP/14 fix log_write() format string regression from TF/03.
554 Bugzilla 1152. Patch from Dmitry Isaikin.
560 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
562 PP/02 Harmonised TLS library version reporting; only show if debugging.
563 Layout now matches that introduced for other libraries in 4.74 PP/03.
565 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
567 PP/04 New "dns_use_edns0" global option.
569 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
572 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
573 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
575 TK/01 Updated PolarSSL code to 0.14.2.
576 Bugzilla 1097. Patch from Andreas Metzler.
578 PP/07 Catch divide-by-zero in ${eval:...}.
581 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
584 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
585 format-string attack -- SECURITY: remote arbitrary code execution.
587 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
588 time unintentionally subject to list matching rules, letting the header
589 cause arbitrary Exim lookups (of items which can occur in lists, *not*
590 arbitrary string expansion). This allowed for information disclosure.
592 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
593 INT_MIN/-1 -- value coerced to INT_MAX.
599 NM/01 Workround for PCRE version dependency in version reporting
602 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
603 This fixes portability to compilers other than gcc, notably
604 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
606 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
607 makefiles for portability to HP-UX and POSIX correctness.
609 PP/01 Permit LOOKUP_foo enabling on the make command-line.
610 Also via indented variable definition in the Makefile.
611 (Debugging by Oliver Heesakkers).
613 PP/02 Restore caching of spamd results with expanded spamd_address.
614 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
616 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
617 Improves build reliability. Fix from: Frank Elsner
619 NM/02 Fix wide character breakage in the rfc2047 coding
620 Fixes bug 1064. Patch from Andrey N. Oktyabrski
622 NM/03 Allow underscore in dnslist lookups
623 Fixes bug 1026. Patch from Graeme Fowler
625 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
626 Code patches from Adam Ciarcinski of NetBSD.
628 NM/04 Fixed exiqgrep to cope with mailq missing size issue
631 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
632 is logged, to avoid truncation. Patch from John Horne.
634 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
635 Patch from Jakob Hirsch.
637 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
638 SQL string expansion failure details.
639 Patch from Andrey Oktyabrski.
641 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
642 Patch from Simon Arlott.
644 PP/09 New lookups functionality failed to compile on old gcc which rejects
645 extern declarations in function scope.
646 Patch from Oliver Fleischmann
648 PP/10 Use sig_atomic_t for flags set from signal handlers.
649 Check getgroups() return and improve debugging.
650 Fixed developed for diagnosis in bug 927 (which turned out to be
653 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
654 Patch from Mark Zealey.
656 PP/12 Bugzilla 1056: Improved spamd server selection.
657 Patch from Mark Zealey.
659 PP/13 Bugzilla 1086: Deal with maildir quota file races.
660 Based on patch from Heiko Schlittermann.
662 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
663 Patch from Uwe Doering, sign-off by Michael Haardt.
665 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
666 variable declaration deep within a block. Bug and patch from
669 PP/15 lookups-Makefile IRIX compatibilty coercion.
671 PP/16 Make DISABLE_DKIM build knob functional.
673 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
674 Patch by Simon Arlott
676 TF/03 Fix valgrind.h portability to C89 compilers that do not support
677 variable argument macros. Our copy now differs from upstream.
683 TF/01 Failure to get a lock on a hints database can have serious
684 consequences so log it to the panic log.
686 TF/02 Log LMTP confirmation messages in the same way as SMTP,
687 controlled using the smtp_confirmation log selector.
689 TF/03 Include the error message when we fail to unlink a spool file.
691 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
692 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
693 for maintaining out-of-tree patches for some time.
695 PP/01 Bugzilla 139: Documentation and portability issues.
696 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
697 Handle per-OS dynamic-module compilation flags.
699 PP/02 Let /dev/null have normal permissions.
700 The 4.73 fixes were a little too stringent and complained about the
701 permissions on /dev/null. Exempt it from some checks.
702 Reported by Andreas M. Kirchwitz.
704 PP/03 Report version information for many libraries, including
705 Exim version information for dynamically loaded libraries. Created
706 version.h, now support a version extension string for distributors
707 who patch heavily. Dynamic module ABI change.
709 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
710 privilege escalation vulnerability whereby the Exim run-time user
711 can cause root to append content of the attacker's choosing to
714 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
717 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
718 If dropping privileges for untrusted macros, we disabled normal logging
719 on the basis that it would fail; for the Exim run-time user, this is not
720 the case, and it resulted in successful deliveries going unlogged.
721 Fixed. Reported by Andreas Metzler.
727 PP/01 Date: & Message-Id: revert to normally being appended to a message,
728 only prepend for the Resent-* case. Fixes regression introduced in
729 Exim 4.70 by NM/22 for Bugzilla 607.
731 PP/02 Include check_rfc2047_length in configure.default because we're seeing
732 increasing numbers of administrators be bitten by this.
734 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
736 PP/03 Bugzilla 994: added openssl_options main configuration option.
738 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
740 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
742 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
744 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
745 without a peer certificate, leading to a segfault because of an
746 assumption that peers always have certificates. Be a little more
747 paranoid. Problem reported by Martin Tscholak.
749 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
750 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
751 NB: ClamAV planning to remove STREAM in "middle of 2010".
752 CL also introduces -bmalware, various -d+acl logging additions and
753 more caution in buffer sizes.
755 PP/09 Implemented reverse_ip expansion operator.
757 PP/10 Bugzilla 937: provide a "debug" ACL control.
759 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
761 PP/12 Bugzilla 973: Implement --version.
763 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
765 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
767 PP/15 Bugzilla 816: support multiple condition rules on Routers.
769 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
770 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
771 ignore trailing whitespace.
773 JJ/02 prevent non-panic DKIM error from being sent to paniclog
775 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
778 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
779 Notification from Dr Andrew Aitchison.
781 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
782 ExtendedDetectionInfo response format.
783 Notification from John Horne.
785 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
788 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
789 XSL and documented dependency on system catalogs, with examples of how
792 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
795 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
796 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
797 configuration file which is writeable by the Exim user or group.
799 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
800 of configuration files to cover files specified with the -C option if
801 they are going to be used with root privileges, not just the default
804 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
805 option (effectively making it always true).
807 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
808 files to be used while preserving root privileges.
810 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
811 that rogue child processes cannot use them.
813 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
814 run-time user, instead of root.
816 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
817 Exim run-time user without dropping privileges.
819 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
820 result string, instead of calling string_vformat() twice with the same
823 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
824 for other users. Others should always drop root privileges if they use
825 -C on the command line, even for a whitelisted configure file.
827 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
829 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
835 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
836 $data_path, and $header_path variables; fixed documentation bugs and
839 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
840 exipick to access non-standard spools, including the "frozen" queue
843 NM/01 Bugzilla 965: Support mysql stored procedures.
844 Patch from Alain Williams
846 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
848 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
849 Patch from Andreas Metzler
851 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
852 Patch from Kirill Miazine
854 NM/05 Bugzilla 671: Added umask to procmail example.
856 JJ/03 installed exipick 20100323.0, fixing doc bug
858 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
859 directory. Notification and patch from Dan Rosenberg.
861 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
863 TK/02 Improve log output when DKIM signing operation fails.
865 MH/01 Treat the transport option dkim_domain as a colon separated
866 list, not as a single string, and sign the message with each element,
867 omitting multiple occurences of the same signer.
869 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
870 Bugzilla 985, 986. Patch by Simon Arlott
872 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
873 Patch by Simon Arlott
875 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
876 MBX locking. Notification from Dan Rosenberg.
882 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
884 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
886 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
888 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
890 NM/04 Bugzilla 915: Flag broken perl installation during build.
896 TK/01 Added patch by Johannes Berg that expands the main option
897 "spamd_address" if it starts with a dollar sign.
899 TK/02 Write list of recipients to X-Envelope-Sender header when building
900 the mbox-format spool file for content scanning (suggested by Jakob
903 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
904 (http://www.dcc-servers.net/) support via dccifd. Activated by
905 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
907 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
908 by Mark Daniel Reidel <mr@df.eu>.
910 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
911 When building exim an external PCRE library is now needed -
912 PCRE is a system library on the majority of modern systems.
913 See entry on PCRE_LIBS in EDITME file.
915 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
916 conversation. Added nologin parameter to request.
917 Patch contributed by Kirill Miazine.
919 TF/01 Do not log submission mode rewrites if they do not change the address.
921 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
923 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
924 log files in place. Contributed by Roberto Lima.
926 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
928 TF/03 Bugzilla 615: When checking the local_parts router precondition
929 after a local_part_suffix or local_part_prefix option, Exim now
930 does not use the address's named list lookup cache, since this
931 contains cached lookups for the whole local part.
933 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
934 Robert Millan. Documentation is in experimental-spec.txt.
936 TF/04 Bugzilla 668: Fix parallel build (make -j).
938 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
940 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
941 Patch provided by Jan Srzednicki.
943 TF/05 Leading white space used to be stripped from $spam_report which
944 wrecked the formatting. Now it is preserved.
946 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
947 that they are available at delivery time.
949 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
951 TF/08 TLS error reporting now respects the incoming_interface and
952 incoming_port log selectors.
954 TF/09 Produce a more useful error message if an SMTP transport's hosts
955 setting expands to an empty string.
957 NM/06 Bugzilla 744: EXPN did not work under TLS.
958 Patch provided by Phil Pennock.
960 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
961 Patch provided by Richard Godbee.
963 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
964 acl_smtp_notquit, added index entry.
966 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
967 Patch provided by Eugene Bujak.
969 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
970 accept(). Patch provided by Maxim Dounin.
972 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
973 Patch provided by Phil Pennock.
975 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
977 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
978 Patch provided by Brad "anomie" Jorsch.
980 NM/14 Bugzilla 622: Added timeout setting to transport filter.
981 Patch provided by Dean Brooks.
983 TK/05 Add native DKIM support (does not depend on external libraries).
985 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
986 Patch provided by Graeme Fowler.
988 NM/16 Bugzilla 851: Documentation example syntax fix.
990 NM/17 Changed NOTICE file to remove references to embedded PCRE.
992 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
995 NM/19 Bugzilla 745: TLS version reporting.
996 Patch provided by Phil Pennock.
998 NM/20 Bugzilla 167: bool: condition support.
999 Patch provided by Phil Pennock.
1001 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1002 clients. Patch provided by Phil Pennock.
1004 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1005 Patch provided by Brad "anomie" Jorsch.
1007 NM/23 Bugzilla 687: Fix misparses in eximstats.
1008 Patch provided by Heiko Schlittermann.
1010 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1011 Patch provided by Heiko Schlittermann.
1013 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1014 plus update to original patch.
1016 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1018 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1019 Patch provided by David Brownlee.
1021 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1023 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1025 NM/30 Bugzilla 888: TLS documentation bugfixes.
1027 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1029 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1030 Unlike the original bugzilla I have changed all shell scripts in src tree.
1032 NM/33 Bugzilla 898: Transport filter timeout fix.
1033 Patch by Todd Rinaldo.
1035 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1036 Patch by Serge Demonchaux.
1038 NM/35 Bugzilla 39: Base64 decode bug fixes.
1039 Patch by Jakob Hirsch.
1041 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1043 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1045 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1047 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1053 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1054 ALT-N's libdkim that I have put here:
1055 http://duncanthrax.net/exim-experimental/
1057 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1058 were called 'true' and it seems that C99 defines that as a reserved
1059 keyword to be used with 'bool' variable types. That means you could
1060 not include C99-style headers which use bools without triggering
1061 build errors in sieve.c.
1063 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1064 as mailq or other aliases. Changed the --help handling significantly
1065 to do whats expected. exim_usage() emits usage/help information.
1067 SC/01 Added the -bylocaldomain option to eximstats.
1069 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1071 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1073 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1079 PH/01 Another patch from the Sieve maintainer.
1081 PH/02 When an IPv6 address is converted to a string for single-key lookup
1082 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1083 dots are used instead of colons so that keys in lsearch files need not
1084 contain colons. This was done some time before quoting was made available
1085 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1086 (notated using the quote facility) so as to distinguish them from IPv4
1087 keys. This meant that lookups for IP addresses in host lists did not work
1088 for iplsearch lookups.
1090 This has been fixed by arranging for IPv6 addresses to be expressed with
1091 colons if the lookup type is iplsearch. This is not incompatible, because
1092 previously such lookups could never work.
1094 The situation is now rather anomolous, since one *can* have colons in
1095 ordinary lsearch keys. However, making the change in all cases is
1096 incompatible and would probably break a number of configurations.
1098 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1101 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1102 conversion specification without a maximum field width, thereby enabling
1103 a rogue spamd server to cause a buffer overflow. While nobody in their
1104 right mind would setup Exim to query an untrusted spamd server, an
1105 attacker that gains access to a server running spamd could potentially
1106 exploit this vulnerability to run arbitrary code as the Exim user.
1108 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1109 $primary_hostname instead of what libspf2 thinks the hosts name is.
1111 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1112 a directory entry by the name of the lookup key. Previously, if a
1113 symlink pointed to a non-existing file or a file in a directory that
1114 Exim lacked permissions to read, a lookup for a key matching that
1115 symlink would fail. Now it is enough that a matching directory entry
1116 exists, symlink or not. (Bugzilla 503.)
1118 PH/03 The body_linecount and body_zerocount variables are now exported in the
1121 PH/04 Added the $dnslist_matched variable.
1123 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1124 This means they are set thereafter only if the connection becomes
1127 PH/06 Added the client_condition to authenticators so that some can be skipped
1128 by clients under certain conditions.
1130 PH/07 The error message for a badly-placed control=no_multiline_responses left
1131 "_responses" off the end of the name.
1133 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1135 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1136 (without spaces) instead of just copying the configuration text.
1138 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1140 PH/11 Added $max_received_linelength.
1142 PH/12 Added +ignore_defer and +include_defer to host lists.
1144 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1145 way in which PCRE > 7.0 is built.
1147 PH/14 Implemented queue_only_load_latch.
1149 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1150 MAIL command. The effect was to mangle the value on 64-bit systems.
1152 PH/16 Another patch from the Sieve maintainer.
1154 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1156 PH/18 If a system quota error occurred while trying to create the file for
1157 a maildir delivery, the message "Mailbox is full" was not appended to the
1158 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1159 only to a quota excession during the actual writing of the file.
1161 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1162 characters?) which causes problems in log lines. The DN values are now
1163 passed through string_printing() before being added to log lines.
1165 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1166 and InterBase are left for another time.)
1168 PH/21 Added message_body_newlines option.
1170 PH/22 Guard against possible overflow in moan_check_errorcopy().
1172 PH/23 POSIX allows open() to be a macro; guard against that.
1174 PH/24 If the recipient of an error message contained an @ in the local part
1175 (suitably quoted, of course), incorrect values were put in $domain and
1176 $local_part during the evaluation of errors_copy.
1182 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1183 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1186 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1187 issue a MAIL command.
1189 PH/02 In an ACL statement such as
1191 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1193 if a client was not listed at all, or was listed with a value other than
1194 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1195 the condition was not true (as it should be), so access was not denied.
1196 The bug was that the ! inversion was incorrectly passed on to the second
1197 item. This has been fixed.
1199 PH/03 Added additional dnslists conditions == and =& which are different from
1200 = and & when the dns lookup returns more than one IP address.
1202 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1203 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1205 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1206 FSYNC, which compiles an option called disable_fsync that allows for
1207 bypassing fsync(). The documentation is heavily laced with warnings.
1209 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1211 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1212 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1213 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1214 including adding "make clean"; (3) Added -fPIC when compiling the test
1215 dynamically loaded module, to get rid of a warning.
1217 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1218 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1219 = 0s. The bug is otherwise harmless.
1221 PH/07 There was a bug in the dovecot authenticator such that the value of
1222 $auth1 could be overwritten, and so not correctly preserved, after a
1223 successful authentication. This usually meant that the value preserved by
1224 the server_setid option was incorrect.
1226 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1228 PH/09 Installed PCRE release 7.0.
1230 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1231 run for batched SMTP input. It is now run at the start of every message
1232 in the batch. While fixing this I discovered that the process information
1233 (output by running exiwhat) was not always getting set for -bs and -bS
1234 input. This is fixed, and it now also says "batched" for BSMTP.
1236 PH/11 Added control=no_pipelining.
1238 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1239 patch, slightly modified), and move the expansion of helo_data till after
1240 the connection is made in the smtp transport (so it can use these
1243 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1245 PH/14 Added log_selector = +pid.
1247 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1249 PH/16 Add ${if forany and ${if forall.
1251 PH/17 Added dsn_from option to vary the From: line in DSNs.
1253 PH/18 Flush SMTP output before performing a callout, unless control =
1254 no_callout_flush is set.
1256 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1257 was true (the default) a successful delivery failed to delete the retry
1258 item, thus causing premature timeout of the address. The bug is now
1261 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1263 PH/21 Long custom messages for fakedefer and fakereject are now split up
1264 into multiline reponses in the same way that messages for "deny" and
1265 other ACL rejections are.
1267 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1268 with slight modification.
1270 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1271 draft, changing the syntax and factoring some duplicate code".
1273 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1274 for deliveries of the second and subsequent messages over the same SMTP
1277 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1278 ${reduce, with only minor "tidies".
1280 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1282 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1283 expansion side effects.
1285 PH/27 When a message times out after an over-quota error from an Exim-imposed
1286 quota, the bounce message says "mailbox is full". This message was not
1287 being given when it was a system quota that was exceeded. It now should
1290 MH/03 Made $recipients available in local_scan(). local_scan() already has
1291 better access to the recipient list through recipients_list[], but
1292 $recipients can be useful in postmaster-provided expansion strings.
1294 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1295 in the case of a MAIL command with additional options following the
1296 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1297 were accidentally chopped off.
1299 PH/29 SMTP synchronization checks are implemented when a command is read -
1300 there is a check that no more input is waiting when there shouldn't be
1301 any. However, for some commands, a delay in an ACL can mean that it is
1302 some time before the response is written. In this time, more input might
1303 arrive, invalidly. So now there are extra checks after an ACL has run for
1304 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1305 pipelining has not been advertised.
1307 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1309 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1310 returned in the SMTP response when smtp_return_error_details was false.
1311 This has been fixed.
1313 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1314 instead of the C I/O that was originally supplied, because problems were
1315 reported on Solaris.
1317 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1318 Exim which did not show up earlier: it was assuming that a call to
1319 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1320 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1321 was a macro that became an assignment, so it seemed to work. This has
1322 changed to a proper function call with a void return, hence the compile
1323 error. Exim's code has been fixed.
1325 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1328 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1330 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1332 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1333 because it turns out to be too big or there is a timeout) there is no
1334 "Completed" line in the log. When some messages of this type were
1335 selected by exigrep, they were listed as "not completed". Others were
1336 picked up by some special patterns. I have improved the selection
1337 criteria to be more general.
1339 PH/38 The host_find_failed option in the manualroute router can now be set
1340 to "ignore", to completely ignore a host whose IP address cannot be
1341 found. If all hosts are ignored, the behaviour is controlled by the new
1342 host_all_ignored option.
1344 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1345 homing or because of multiple MX records with /mx) generated more than
1346 one IP address, and the following item turned out to be the local host,
1347 all the secondary addresses of the first item were incorrectly removed
1348 from the list, along with the local host and any following hosts (which
1349 is what is supposed to happen).
1351 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1352 whoever called Exim into the -H file. In the case of the daemon it was
1353 behaving confusingly. When first started, it used values for whoever
1354 started the daemon, but after a SIGHUP it used the Exim user (because it
1355 calls itself on a restart). I have changed the code so that it now always
1358 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1359 message are rejected with the same error (e.g. no authentication or bad
1360 sender address), and a DATA command is nevertheless sent (as can happen
1361 with PIPELINING or a stupid MUA), the error message that was given to the
1362 RCPT commands is included in the rejection of the DATA command. This is
1363 intended to be helpful for MUAs that show only the final error to their
1366 PH/42 Another patch from the Sieve maintainer.
1368 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1369 Eximstats - Fixed some broken HTML links and added missing column headers
1371 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1372 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1374 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1376 PH/43 Yet another patch from the Sieve maintainer.
1378 PH/44 I found a way to check for a TCP/IP connection going away before sending
1379 the response to the final '.' that terminates a message, but only in the
1380 case where the client has not sent further data following the '.'
1381 (unfortunately, this is allowed). However, in many cases there won't be
1382 any further data because there won't be any more messages to send. A call
1383 to select() can be used: if it shows that the input is "ready", there is
1384 either input waiting, or the socket has been closed. An attempt to read
1385 the next input character can distinguish the two cases. Previously, Exim
1386 would have sent an OK response which the client would never have see.
1387 This could lead to message repetition. This fix should cure that, at
1388 least in a lot of common cases.
1390 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1391 advertised in response to EHLO.
1397 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1398 fixed by 4.65/MH/01 (is this a record?) are fixed:
1400 (i) An empty string was always treated as zero by the numeric comparison
1401 operators. This behaviour has been restored.
1403 (ii) It is documented that the numeric comparison operators always treat
1404 their arguments as decimal numbers. This was broken in that numbers
1405 starting with 0 were being interpreted as octal.
1407 While fixing these problems I realized that there was another issue that
1408 hadn't been noticed. Values of message_size_limit (both the global option
1409 and the transport option) were treated as octal if they started with 0.
1410 The documentation was vague. These values are now always treated as
1411 decimal, and I will make that clear in the documentation.
1417 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1418 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1421 MH/01 Don't check that the operands of numeric comparison operators are
1422 integers when their expansion is in "skipping" mode (fixes bug
1423 introduced by 4.64-PH/07).
1425 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1426 child addresses, Exim now panics and dies. Previously, because the count
1427 is held in a short int, deliveries were likely to be lost. As such a
1428 large number of recipients for a single message is ridiculous
1429 (performance will be very, very poor), I have chosen to impose a limit
1430 rather than extend the field.
1436 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1437 leftover -K file (the existence of which was triggered by #402).
1438 While we were at it, introduced process PID as part of the -K
1439 filename. This should rule out race conditions when creating
1442 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1443 processing considerably. Previous code took too long for large mails,
1444 triggering a timeout which in turn triggers #401.
1446 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1447 in the DK code in transports.c. sendfile() is not really portable,
1448 hence the _LINUX specificness.
1450 TF/01 In the add_headers option to the mail command in an Exim filter,
1451 there was a bug that Exim would claim a syntax error in any
1452 header after the first one which had an odd number of characters
1455 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1456 callout verification, Exim cached a "reject" for the entire domain. This
1457 is correct for most verifications, but it is not correct for a recipient
1458 verification with use_sender or use_postmaster set, because in that case
1459 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1460 case of MAIL FROM:<> rejection from other early rejections (e.g.
1461 rejection of HELO). When verifying a recipient using a non-null MAIL
1462 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1463 Whatever the result of the callout, the value of the domain cache is
1464 left unchanged (for any other kind of callout, getting as far as trying
1465 RCPT means that the domain itself is ok).
1467 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
1470 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
1471 manifest itself as EPIPE rather than ECONNECT. When tidying away a
1472 session, the daemon ignores ECONNECT errors and logs others; it now
1473 ignores EPIPE as well.
1475 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
1476 (quoted-printable decoding).
1478 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
1479 later the small subsequent patch to fix an introduced bug.
1481 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
1483 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
1485 PH/08 An error is now given if message_size_limit is specified negative.
1487 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
1488 to be given (somewhat) arbitrary names.
1490 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
1493 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
1494 miscellaneous code fixes
1496 PH/10 Added the log_reject_target ACL modifier to specify where to log
1499 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
1500 hostname. This is wrong, because it relates to the incoming message (and
1501 probably the interface on which it is arriving) and not to the outgoing
1502 callout (which could be using a different interface). This has been
1503 changed to use the value of the helo_data option from the smtp transport
1504 instead - this is what is used when a message is actually being sent. If
1505 there is no remote transport (possible with a router that sets up host
1506 addresses), $smtp_active_hostname is used.
1508 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
1509 tweaks were necessary in order to get it to work (see also 21 below):
1510 (a) The code assumed that strncpy() returns a negative number on buffer
1511 overflow, which isn't the case. Replaced with Exim's string_format()
1513 (b) There were several signed/unsigned issues. I just did the minimum
1514 hacking in of casts. There is scope for a larger refactoring.
1515 (c) The code used strcasecmp() which is not a standard C function.
1516 Replaced with Exim's strcmpic() function.
1517 (d) The code set only $1; it now sets $auth1 as well.
1518 (e) A simple test gave the error "authentication client didn't specify
1519 service in request". It would seem that Dovecot has changed its
1520 interface. Fortunately there's a specification; I followed it and
1521 changed what the client sends and it appears to be working now.
1523 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
1526 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
1527 address A is aliased to B and C, where B exists and C does not. Without
1528 -v the output is "A verified" because verification stops after a
1529 successful redirection if more than one address is generated. However,
1530 with -v the child addresses are also verified. Exim was outputting "A
1531 failed to verify" and then showing the successful verification for C,
1532 with its parentage. It now outputs "B failed to verify", showing B's
1533 parentage before showing the successful verification of C.
1535 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
1536 look up a TXT record in a specific list after matching in a combined
1539 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
1540 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
1541 they consult the DNS. I had assumed they would set it the way they
1542 wanted; and indeed my experiments on Linux seem to show that in some
1543 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
1544 To be on the safe side, however, I have now made the interface to
1545 host_find_byname() similar to host_find_bydns(), with an argument
1546 containing the DNS resolver options. The host_find_byname() function now
1547 sets these options at its start, just as host_find_bydns() does. The smtp
1548 transport options dns_qualify_single and dns_search_parents are passed to
1549 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
1550 of host_find_byname() use the default settings of RES_DEFNAMES
1551 (qualify_single) but not RES_DNSRCH (search_parents).
1553 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
1554 spool_read_header() do less string testing, by means of a preliminary
1555 switch on the second character of optional "-foo" lines. (This is
1556 overdue, caused by the large number of possibilities that now exist.
1557 Originally there were few.) While I was there, I also converted the
1558 str(n)cmp tests so they don't re-test the leading "-" and the first
1559 character, in the hope this might squeeze out yet more improvement.
1561 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
1562 flag allowing group syntax was set by the header_syntax check but not
1563 turned off, possible causing trouble later; (2) The flag was not being
1564 set at all for the header_verify test, causing "group"-style headers to
1565 be rejected. I have now set it in this case, and also caused header_
1566 verify to ignore an empty address taken from a group. While doing this, I
1567 came across some other cases where the code for allowing group syntax
1568 while scanning a header line wasn't quite right (mostly, not resetting
1569 the flag correctly in the right place). These bugs could have caused
1570 trouble for malformed header lines. I hope it is now all correct.
1572 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
1573 with the "reply" argument non-NULL. The code, however (which originally
1574 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
1575 but it didn't always do it. This confused somebody who was copying the
1576 code for some other use. I have removed all the tests.
1578 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
1579 feature that was used to support insecure browsers during the U.S. crypto
1580 embargo. It requires special client support, and Exim is probably the
1581 only MTA that supported it -- and would never use it because real RSA is
1582 always available. This code has been removed, because it had the bad
1583 effect of slowing Exim down by computing (never used) parameters for the
1584 RSA_EXPORT functionality.
1586 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
1587 authenticator to fail if there's a tab character in the incoming data
1588 (there should never be unless someone is messing about, as it's supposed
1589 to be base64-encoded). Also added, on Timo's advice, the "secured" option
1590 if the connection is using TLS or if the remote IP is the same as the
1591 local IP, and the "valid-client-cert option" if a client certificate has
1594 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
1595 authenticators. This can be used for authorization after authentication
1596 succeeds. (In the case of plaintext, it servers for both authentication
1599 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
1600 if any retry times were supplied.
1602 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
1603 connection, where there is no client IP address to check. In this
1604 situation, the verify now always succeeds.
1606 PH/25 Applied John Jetmore's -Mset patch.
1608 PH/26 Added -bem to be like -Mset, but loading a message from a file.
1610 PH/27 In a string expansion for a processed (not raw) header when multiple
1611 headers of the same name were present, leading whitespace was being
1612 removed from all of them, but trailing whitespace was being removed only
1613 from the last one. Now trailing whitespace is removed from each header
1614 before concatenation. Completely empty headers in a concatenation (as
1615 before) are ignored.
1617 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
1618 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
1620 PH/29 [Removed. This was a change that I later backed out, and forgot to
1621 correct the ChangeLog entry (that I had efficiently created) before
1622 committing the later change.]
1624 PH/30 Exim was sometimes attempting to deliver messages that had suffered
1625 address errors (4xx response to RCPT) over the same connection as other
1626 messages routed to the same hosts. Such deliveries are always "forced",
1627 so retry times are not inspected. This resulted in far too many retries
1628 for the affected addresses. The effect occurred only when there were more
1629 hosts than the hosts_max_try setting in the smtp transport when it had
1630 the 4xx errors. Those hosts that it had tried were not added to the list
1631 of hosts for which the message was waiting, so if all were tried, there
1632 was no problem. Two fixes have been applied:
1634 (i) If there are any address or message errors in an SMTP delivery, none
1635 of the hosts (tried or untried) are now added to the list of hosts
1636 for which the message is waiting, so the message should not be a
1637 candidate for sending over the same connection that was used for a
1638 successful delivery of some other message. This seems entirely
1639 reasonable: after all the message is NOT "waiting for some host".
1640 This is so "obvious" that I'm not sure why it wasn't done
1641 previously. Hope I haven't missed anything, but it can't do any
1642 harm, as the worst effect is to miss an optimization.
1644 (ii) If, despite (i), such a delivery is accidentally attempted, the
1645 routing retry time is respected, so at least it doesn't keep
1646 hammering the server.
1648 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
1649 in ${readsocket because some servers need this prod.
1651 PH/32 Added some extra debug output when updating a wait-xxx database.
1653 PH/33 The hint "could be header name not terminated by colon", which has been
1654 given for certain expansion errors for a long time, was not being given
1655 for the ${if def:h_colon_omitted{... case.
1657 PH/34 The spec says: "With one important exception, whenever a domain list is
1658 being scanned, $domain contains the subject domain." There was at least
1659 one case where this was not true.
1661 PH/35 The error "getsockname() failed: connection reset by peer" was being
1662 written to the panic log as well as the main log, but it isn't really
1663 panic-worthy as it just means the connection died rather early on. I have
1664 removed the panic log writing for the ECONNRESET error when getsockname()
1667 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
1668 runs only) independently of the message's sender address. This meant
1669 that, if the 4xx error was in fact related to the sender, a different
1670 message to the same recipient with a different sender could confuse
1671 things. In particualar, this can happen when sending to a greylisting
1672 server, but other circumstances could also provoke similar problems.
1673 I have changed the default so that the retry time for these errors is now
1674 based a combination of the sender and recipient addresses. This change
1675 can be overridden by setting address_retry_include_sender=false in the
1678 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
1679 remote server are returned as part of bounce messages. This was not
1680 happening for LMTP over a pipe (the lmtp transport), but now it is the
1681 same for both kinds of LMTP.
1683 PH/38 Despite being documented as not happening, Exim was rewriting addresses
1684 in header lines that were in fact CNAMEs. This is no longer the case.
1686 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
1687 and queue runs started by the daemon processed all messages. This has
1688 been fixed so that -R and -S can now usefully be given with -q<time>.
1690 PH/40 Import PCRE release 6.7 (fixes some bugs).
1692 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
1694 PH/42 Give an error if -q is specified more than once.
1696 PH/43 Renamed the variables $interface_address and $interface_port as
1697 $received_ip_address and $received_port, to make it clear that these
1698 values apply to message reception, and not to the outgoing interface when
1699 a message is delivered. (The old names remain recognized, of course.)
1701 PH/44 There was no timeout on the connect() call when using a Unix domain
1702 socket in the ${readsocket expansion. There now is.
1704 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
1705 be meaningful with "accept".
1707 SC/01 Eximstats V1.43
1708 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
1710 SC/02 Eximstats V1.44
1711 Use a glob alias rather than an array ref in the generated
1712 parser. This improves both readability and performance.
1714 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
1715 Collect SpamAssassin and rejection statistics.
1716 Don't display local sender or destination tables unless
1717 there is data to show.
1718 Added average volumes into the top table text output.
1720 SC/04 Eximstats V1.46
1721 Collect data on the number of addresses (recipients)
1722 as well as the number of messages.
1724 SC/05 Eximstats V1.47
1725 Added 'Message too big' to the list of mail rejection
1726 reasons (thanks to Marco Gaiarin).
1728 SC/06 Eximstats V1.48
1729 Mainlog lines which have GMT offsets and are too short to
1730 have a flag are now skipped.
1732 SC/07 Eximstats V1.49 (Alain Williams)
1733 Added the -emptyok flag.
1735 SC/08 Eximstats V1.50
1736 Fixes for obtaining the IP address from reject messages.
1738 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
1739 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
1740 whitesspace changes from 4.64-PH/27
1742 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
1745 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
1746 are found, allow negative numbers in numeric criteria)
1748 JJ/06 exipick.20061117.2, added new $message_body_missing variable
1750 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
1751 to match changes made in 4.64-PH/43
1753 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
1755 PH/47 Put in an explicit test for a DNS lookup of an address record where the
1756 "domain" is actually an IP address, and force a failure. This locks out
1757 those revolvers/nameservers that support "A-for-A" lookups, in
1758 contravention of the specifications.
1760 PH/48 When a host name was looked up from an IP address, and the subsequent
1761 forward lookup of the name timed out, the host name was left in
1762 $sender_host_name, contrary to the specification.
1764 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
1765 restricted to single-key lookups, Exim was not diagnosing an error if
1766 * or *@ was used with a query-style lookup.
1768 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
1770 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
1771 long ago, but noone interested enough thought of it. Let's just say that
1772 the "1.1" means that there are some new functions that weren't there at
1773 some point in the past.
1775 PH/51 Error processing for expansion failure of helo_data from an smtp
1776 transport during callout processing was broken.
1778 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
1779 tested/used via the -bh/-bhc/-bs options.
1781 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
1782 bug, fixed in subsequent PCRE releases).
1784 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
1785 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
1787 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
1793 SC/01 Use a glob alias rather than an array ref in eximstats generated
1794 parser. This improves both readability and performance.
1796 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
1797 Don't display local sender or destination tables in eximstats unless
1798 there is data to show.
1799 Added average volumes into the eximstats top table text output.
1801 SC/03 Collect data on the number of addresses (recipients) as well
1802 as the number of messages in eximstats.
1804 TF/01 Correct an error in the documentation for the redirect router. Exim
1805 does (usually) call initgroups() when daemonizing.
1807 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
1808 with consistent privilege compared to when running as a daemon.
1810 TF/03 Note in the spec that $authenticated_id is not set for local
1811 submissions from trusted users.
1813 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
1814 Thanks to Dean Brooks <dean@iglou.com> for the patch.
1816 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
1817 by adding some example configuration directives to the default
1818 configuration file. A little bit of work is required to uncomment the
1819 directives and define how usernames and passwords are checked, but
1820 there is now a framework to start from.
1822 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
1823 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
1824 without this. I don't know how relevant this is to other LDAP libraries.
1826 PH/02 Add the verb name to the "unknown ACL verb" error.
1828 PH/03 Magnus Holmgren's patch for filter_prepend_home.
1830 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
1832 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
1833 directory not expanded when it should be if an expanded home directory
1834 was set for the address (which is overridden by the transport).
1836 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
1839 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
1840 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
1841 because it is too late at that time, and has no effect.
1843 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
1844 security issue with \' (bugzilla #107). I could not use the
1845 PQescapeStringConn() function, because it needs a PGconn value as one of
1848 PH/08 When testing addresses using -bt, indicate those final addresses that
1849 are duplicates that would not cause an additional delivery. At least one
1850 person was confused, thinking that -bt output corresponded to deliveries.
1851 (Suppressing duplicates isn't a good idea as you lose the information
1852 about possibly different redirections that led to the duplicates.)
1854 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
1855 systems where poll() doesn't work, in particular OS X.
1857 PH/10 Added more information to debugging output for retry time not reached.
1859 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
1860 operations in malware.c.
1862 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
1865 PH/13 If write_rejectlog was set false when logging was sent to syslog with
1866 syslog_duplication set false, log lines that would normally be written
1867 both the the main log and to the reject log were not written to syslog at
1870 PH/14 In the default configuration, change the use of "message" in ACL warn
1871 statements to "add_header".
1873 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
1874 not followed by a command (e.g. "seen endif").
1876 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
1877 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
1880 PH/17 Added extra conditions to the default value of delay_warning_condition
1884 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
1885 { match{$h_precedence:}{(?i)bulk|list|junk} } \
1886 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
1889 The Auto-Submitted: and various List- headers are standardised, whereas I
1890 don't think Precedence: ever was.
1892 PH/18 Refactored debugging code in route_finduser() to show more information,
1893 in particular, the error code if getpwnam() issues one.
1895 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
1896 This is apparently needed in addition to the PH/07 change above to avoid
1897 any possible encoding problems.
1899 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
1900 but not after initializing Perl.
1902 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
1903 output them only if debugging. By default they are written stderr,
1904 apparently, which is not desirable.
1906 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
1909 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
1912 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1914 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1915 authenticated or an ident call has been made. Suppress the default
1916 values for $authenticated_id and $authenticated_sender (but permit -oMai
1917 and -oMas) when testing with -bh.
1919 PH/24 Re-jigged the order of the tests in the default configuration so that the
1920 tests for valid domains and recipients precede the DNS black list and CSA
1921 tests, on the grounds that those ones are more expensive.
1923 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1924 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1925 This bug exists in every version of Exim that I still have, right back to
1928 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1929 However, an attempt to turn on case-sensitivity in a regex key by
1930 including (?-i) didn't work because the subject string was already
1931 lowercased, and the effects were non-intuitive. It turns out that a
1932 one-line patch can be used to allow (?-i) to work as expected.
1938 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1939 other effects) broke the use of negated acl sub-conditions.
1941 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1944 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1945 "Deny" causes Exim to reject the incoming connection with a 554 error.
1946 Unfortunately, if there is a major crisis, such as a disk failure,
1947 tcp-wrappers gives "deny", whereas what one would like would be some
1948 kind of temporary error. A kludge has been added to help with this.
1949 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1950 554 error is used if errno is still zero or contains ENOENT (which occurs
1951 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1954 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1956 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1957 errors. Otherwise a message that provokes a temporary error (when other
1958 messages do not) can cause a whole host to time out.
1960 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1961 addresses were routed directly to files or pipes from a redirect router.
1962 File deliveries just didn't batch; pipe deliveries might have suffered
1965 PH/06 A failure to get a lock for a hints database would erroneously always say
1966 "Failed to get write lock", even when it was really a read lock.
1968 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1969 of 0600. This has been changed to use the value of the lockfile_mode
1970 option (which defaults to 0600).
1972 PH/08 Applied small patch from the Sieve maintainer.
1974 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1975 folder from quota calculations, a direct delivery into this folder messed
1976 up the contents of the maildirsize file. This was because the regex was
1977 used only to exclude .Trash (or whatever) when the size of the mailbox
1978 was calculated. There was no check that a delivery was happening into an
1979 excluded directory. This bug has been fixed by ignoring all quota
1980 processing for deliveries into excluded directories.
1982 PH/10 Added the maildirfolder_create_regex option to appendfile.
1988 PH/01 The code for finding all the local interface addresses on a FreeBSD
1989 system running IPv6 was broken. This may well have applied to all BSD
1990 systems, as well as to others that have similar system calls. The broken
1991 code found IPv4 interfaces correctly, but gave incorrect values for the
1992 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1993 that it would not match correctly against @[] and not recognize the IPv6
1996 PH/02 The ipliteral router was not recognizing addresses of the form user@
1997 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1999 PH/03 Added disable_ipv6.
2001 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2002 decoded form, because it is most often used to construct To: headers
2003 lines in autoreplies, and the decoded form may well be syntactically
2004 invalid. However, $reply_address has leading white space removed, and all
2005 newlines turned into spaces so that the autoreply transport does not
2008 PH/05 If group was specified without a user on a router, and no group or user
2009 was specified on a transport, the group from the router was ignored.
2011 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2012 for visible compile-time settings that can be used to change these
2013 numbers, for those that want even more. Backwards compatibility with old
2014 spool files has been maintained. However, going back to a previous Exim
2015 release will lost any variables that are in spool files.
2017 PH/07 Two small changes when running in the test harness: increase delay when
2018 passing a TCP/IP connection to a new process, in case the original
2019 process has to generate a bounce, and remove special handling of
2020 127.0.0.2 (sic), which is no longer necessary.
2022 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2023 be the same on different OS.
2025 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2028 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2029 whether --show-vars was specified or not
2031 JJ/02 exipick: Added support for new ACL variable spool format introduced
2034 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2035 syntactically invalid From: or Reply-to: line, and a filter used this to
2036 generate an autoreply, and therefore failed to obtain an address for the
2037 autoreply, Exim could try to deliver to a non-existent relative file
2038 name, causing unrelated and misleading errors. What now happens is that
2039 it logs this as a hard delivery error, but does not attempt to create a
2042 PH/11 The exinext utility has a -C option for testing purposes, but although
2043 the given file was scanned by exinext itself; it wasn't being passed on
2044 when Exim was called.
2046 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2047 an end-of-file indication when reading a command response.
2049 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2050 compiled. In many other places in Exim, IPv6 addresses are always
2051 recognized, so I have changed this. It also means that IPv4 domain
2052 literals of the form [IPV4:n.n.n.n] are now always recognized.
2054 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2055 used if the router is not running as root, for example, when verifying at
2056 ACL time, or when using -bh. The debugging output from this situation was
2057 non-existent - all you got was a failure to exec. I have made two
2060 (a) Failures to set uid/gid, the current directory, or a process leader
2061 in a subprocess such as that created by queryprogram now generate
2062 suitable debugging ouput when -d is set.
2064 (b) The queryprogram router detects when it is not running as root,
2065 outputs suitable debugging information if -d is set, and then runs
2066 the subprocess without attempting to change uid/gid.
2068 PH/15 Minor change to Makefile for building test_host (undocumented testing
2071 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2072 additional section of a DNS packet that returns MX or SRV records.
2073 Instead, it always explicitly searches for A/AAAA records. This avoids
2074 major problems that occur when a DNS server includes only records of one
2075 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2076 fixed another bug: if SRV records were looked up and the corresponding
2077 address records were *not* found in the additional section, the port
2078 values from the SRV records were lost.
2080 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2081 using the correct key (the original address) when searching the retry
2082 rules in order to find which one to use for generating the retry hint.
2084 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2085 adding the default one. Similarly, if it contains a Reply-To: header, the
2086 errors_reply_to option, if set, is not used.
2088 PH/19 When calculating a retry time, Exim used to measure the "time since
2089 failure" by looking at the "first failed" field in the retry record. Now
2090 it does not use this if it is later than than the arrival time of the
2091 message. Instead it uses the arrival time. This makes for better
2092 behaviour in cases where some deliveries succeed, thus re-setting the
2093 "first failed" field. An example is a quota failure for a huge message
2094 when small messages continue to be delivered. Without this change, the
2095 "time since failure" will always be short, possible causing more frequent
2096 delivery attempts for the huge message than are intended.
2097 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2099 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2100 $1, $2, $3) because the numerical variables can be reset during some
2101 expansion items (e.g. "match"), thereby losing the authentication data.
2103 PH/21 Make -bV show the size of off_t variables so that the test suite can
2104 decide whether to run tests for quotas > 2G.
2106 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2107 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2108 filecount value is greater than 2G or if a quota value is greater than 2G
2109 on a system where the size of off_t is not greater than 4, a panic error
2112 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2113 never match. The debug and -bh output now contains an explicit error
2114 message indicating a malformed IPv4 address or mask.
2116 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2117 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2118 PH/23 above applies.
2120 PH/25 Do not write to syslog when running in the test harness. The only
2121 occasion when this arises is a failure to open the main or panic logs
2122 (for which there is an explicit test).
2124 PH/26 Added the /no_tell option to "control=freeze".
2126 PH/27 If a host name lookup failed very early in a connection, for example, if
2127 the IP address matched host_lookup and the reverse lookup yielded a name
2128 that did not have a forward lookup, an error message of the form "no IP
2129 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2130 could be logged. Now it outputs the IP address instead of "NULL".
2132 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2133 allows the sender and the authenticated sender to be set when
2134 submitting a message from within Exim. Since child_open_exim() is
2135 documented for local_scan(), the new function should be too.
2137 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2138 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2139 results in an empty string is now treated as unset.
2141 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2143 PH/31 Added sender_verify_fail logging option.
2145 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2146 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2147 tidied the source and removed it altogether.
2149 PH/33 When a queue run was abandoned because the load average was too high, a
2150 log line was always written; now it is written only if the queue_run log
2151 selector is set. In addition, the log line for abandonment now contains
2152 information about the queue run such as the pid. This is always present
2153 in "start" and "stop" lines but was omitted from the "abandon" line.
2155 PH/34 Omit spaces between a header name and the colon in the error message that
2156 is given when verify = headers_syntax fails (if there are lots of them,
2157 the message gets confusing).
2159 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2160 names, as there are now some PTR records that contain slashes. This check
2161 is only to protect against broken name servers that fall over on strange
2162 characters, so the fact that it applies to all lookups doesn't matter.
2164 PH/36 Now that the new test suite is complete, we can remove some of the
2165 special code in Exim that was needed for the old test suite. For example,
2166 sorting DNS records because real resolvers return them in an arbitrary
2167 order. The new test suite's fake resolver always returns records in the
2170 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2171 bounces) except when queue_only is set, to avoid logging races between
2172 the different processes.
2174 PH/38 Panic-die if .include specifies a non-absolute path.
2176 PH/39 A tweak to the "H" retry rule from its user.
2178 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2179 a label. They prevented compilation on older perls.
2181 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2182 a warning to be raised on newish perls.
2184 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2185 on queue. Changes to match documented behaviour of showing count of
2186 messages matching specified criteria.
2188 PH/40 Changed the default ident timeout from 30s to 5s.
2190 PH/41 Added support for the use of login_cap features, on those BSD systems
2191 that have them, for controlling the resources used by pipe deliveries.
2193 PH/42 The content-scanning code uses fopen() to create files in which to put
2194 message data. Previously it was not paying any attention to the mode of
2195 the files. Exim runs with umask(0) because the rest of the code creates
2196 files with open(), and sets the required mode explicitly. Thus, these
2197 files were ending up world-writeable. This was not a big issue, because,
2198 being within the spool directory, they were not world-accessible. I have
2199 created a function called modefopen, which takes an additional mode
2200 argument. It sets umask(777), creates the file, chmods it to the required
2201 mode, then resets the umask. All the relevant calls to fopen() in the
2202 content scanning code have been changed to use this function.
2204 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2205 to 24 hours. This avoids potential overflow problems when processing G
2206 and H retry rules. I suspect nobody ever tinkers with this value.
2208 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2210 PH/45 When the plaintext authenticator is running as a client, the server's
2211 challenges are checked to ensure they are valid base64 strings. By
2212 default, the authentication attempt is cancelled if an invalid string is
2213 received. Setting client_ignore_invalid_base64 true ignores these errors.
2214 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2215 they are received. Thus, the responses can be made to depend on the
2216 challenges. If an invalid string is ignored, an empty string is placed in
2219 PH/46 Messages that are created by the autoreply transport now contains a
2220 References: header, in accordance with RFCs 2822 and 3834.
2222 PH/47 Added authenticated_sender_force to the smtp transport.
2224 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2226 PH/49 Installed latest patch from the Sieve maintainer.
2228 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2229 was also set, the appendfile transport was unnecessarily scanning a
2230 directory of message files (e.g. for maildir delivery) to find the count
2231 of files (along with the size), even though it did not need this
2232 information. It now does the scan only if it needs to find either the
2233 size of the count of files.
2235 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2237 PH/52 Two bugs concerned with error handling when the smtp transport is
2240 (i) Exim was not creating retry information for temporary errors given
2241 for individual recipients after the DATA command when the smtp transport
2242 was used in LMTP mode. This meant that they could be retried too
2243 frequently, and not timed out correctly.
2245 (ii) Exim was setting the flag that allows error details to be returned
2246 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2247 recipients that were returned after the DATA command.
2249 PH/53 This is related to PH/52, but is more general: for any failing address,
2250 when detailed error information was permitted to be returned to the
2251 sender, but the error was temporary, then after the final timeout, only
2252 "retry timeout exceeded" was returned. Now it returns the full error as
2253 well as "retry timeout exceeded".
2255 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2256 do this, and (what is worse) MTAs that accept it.
2258 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2259 will now be deprecated.
2261 PH/56 New os.c-cygwin from the Cygwin maintainer.
2263 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2264 formats (previously only available in exim formats via -bpr, -bpru,
2265 and -bpra. Now also available in native and exiqgrep formats)
2267 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2268 with very large, slow to parse queues
2270 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2272 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2274 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2275 responses to authentication challenges, though it was showing the
2276 challenges; (ii) I've removed the CR characters from the debug output for
2279 PH/58 Allow for the insertion of a newline as well as a space when a string
2280 is turned into more than one encoded-word during RFC 2047 encoding. The
2281 Sieve code now uses this.
2283 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2284 data_4xx, lost_connection, tls_required.
2286 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2287 message was being sent as an SMTP response.
2289 PH/61 Add -l and -k options to exicyclog.
2291 PH/62 When verifying, if an address was redirected to one new address, so that
2292 verification continued, and the new address failed or deferred after
2293 having set something in $address_data, the value of $address_data was not
2294 passed back to the ACL. This was different to the case when no
2295 redirection occurred. The value is now passed back in both cases.
2297 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2298 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2299 use login_cap.h, so on its own it isn't the distinguishing feature. The
2300 new name refers directly to the setclassresources() function.
2302 PH/65 Added configuration files for NetBSD3.
2304 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2306 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2307 is preferred over IPv4.
2309 PH/68 The bounce_return_message and bounce_return_body options were not being
2310 honoured for bounces generated during the reception of non-SMTP messages.
2311 In particular, this applied to messages rejected by the ACL. This bug has
2312 been fixed. However, if bounce_return_message is true and bounce_return_
2313 body is false, the headers that are returned for a non-SMTP message
2314 include only those that have been read before the error was detected.
2315 (In the case of an ACL rejection, they have all been read.)
2317 PH/69 The HTML version of the specification is now built in a directory called
2318 spec_html instead of spec.html, because the latter looks like a path with
2319 a MIME-type, and this confuses some software.
2321 PH/70 Catch two compiler warnings in sieve.c.
2323 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2324 function verify_get_ident() calls ip_connect() to connect a socket, but
2325 if the "connect()" function timed out, ip_connect() used to close the
2326 socket. However, verify_get_ident() also closes the socket later, and in
2327 between Exim writes to the log, which may get opened at this point. When
2328 the socket was closed in ip_connect(), the log could get the same file
2329 descriptor number as the socket. This naturally causes chaos. The fix is
2330 not to close the socket in ip_connect(); the socket should be closed by
2331 the function that creates it. There was only one place in the code where
2332 this was missing, in the iplookup router, which I don't think anybody now
2333 uses, but I've fixed it anyway.
2335 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2336 well as to direct DNS lookups. Otherwise the handling of names in host
2337 lists is inconsistent and therefore confusing.
2343 PH/01 Two changes to the default runtime configuration:
2345 (1) Move the checks for relay_from_hosts and authenticated clients from
2346 after to before the (commented out) DNS black list checks.
2348 (2) Add control=submission to the relay_from_hosts and authenticated
2349 clients checks, on the grounds that messages accepted by these
2350 statements are most likely to be submissions.
2352 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2354 (1) Generate an error if the third argument for the ${prvs expansion is
2357 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2360 (3) Reset the variables that are obtained from the first argument of
2361 ${prvscheck and used in the second argument before leaving the code,
2362 because their memory is reclaimed, so using them afterwards may do
2365 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2366 one (it's much easier than Tom thought :-).
2368 (5) Because of (4), we can now allow for the use of $prvscheck_result
2369 inside the third argument.
2371 PH/03 For some reason, the default setting of PATH when running a command from
2372 a pipe transport was just "/usr/bin". I have changed it to
2375 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2376 anything to be listed in the output from -bV.
2378 PH/05 When a filter generated an autoreply, the entire To: header line was
2379 quoted in the delivery log line, like this:
2381 => >A.N.Other <ano@some.domain> <original@ddress> ...
2383 This has been changed so that it extracts the operative address. There
2384 may be more than one such address. If so, they are comma-separated, like
2387 => >ano@some.domain,ona@other.domain <original@ddress> ...
2389 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2390 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2391 not being looked up in the rDNS to get a host name, Exim was showing the
2392 IP address twice in Received: lines, even though the IP addresses were
2393 identical. For example:
2395 Received: from [1.2.3.4] (helo=[1.2.3.4])
2397 However, if the real host name was known, it was omitting the HELO data
2398 if it matched the actual IP address. This has been tidied up so that it
2399 doesn't show the same IP address twice.
2401 PH/07 When both +timestamp and +memory debugging was on, the value given by
2402 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2403 called by the debug printing, thereby overwriting the timestamp buffer.
2404 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2406 PH/08 When the original message was included in an autoreply transport, it
2407 always said "this is a copy of the message, including all the headers",
2408 even if body_only or headers_only was set. It now gives an appropriate
2411 PH/09 Applied a patch from the Sieve maintainer which:
2413 o fixes some comments
2414 o adds the (disabled) notify extension core
2415 o adds some debug output for the result of if/elsif tests
2416 o points to the current vacation draft in the documentation
2417 and documents the missing references header update
2421 o fixes a bug in processing the envelope test (when testing
2422 multiple envelope elements, the last element determinted the
2425 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2426 Electronic Mail") by including:
2428 Auto-submitted: auto-generated
2430 in the messages that it generates (bounce messages and others, such as
2431 warnings). In the case of bounce messages for non-SMTP mesages, there was
2432 also a typo: it was using "Auto_submitted" (underscore instead of
2433 hyphen). Since every message generated by Exim is necessarily in response
2434 to another message, thes have all been changed to:
2436 Auto-Submitted: auto-replied
2438 in accordance with these statements in the RFC:
2440 The auto-replied keyword:
2442 - SHOULD be used on messages sent in direct response to another
2443 message by an automatic process,
2445 - MUST NOT be used on manually-generated messages,
2447 - MAY be used on Delivery Status Notifications (DSNs) and Message
2448 Disposition Notifications (MDNs),
2450 - MUST NOT be used on messages generated by automatic or periodic
2451 processes, except for messages which are automatic responses to
2454 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2455 to the default Received: header definition.
2457 PH/12 Added log selector acl_warn_skipped (default on).
2459 PH/13 After a successful wildlsearch lookup, discard the values of numeric
2460 variables because (a) they are in the wrong storage pool and (b) even if
2461 they were copied, it wouldn't work properly because of the caching.
2463 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2464 checking when decoding. Apparently there are clients that generate
2465 overlong encoded strings. Why am I not surprised?
2467 PH/15 If the first argument of "${if match_address" was not empty, but did not
2468 contain an "@" character, Exim crashed. Now it writes a panic log message
2469 and treats the condition as false.
2471 PH/16 In autoreply, treat an empty string for "once" the same as unset.
2473 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
2474 extension "envelope-auth". The code is finished and in agreement with
2475 other implementations, but there is no documentation so far and in fact,
2476 nobody wrote the draft yet. This extension is currently #undef'ed, thus
2477 not changing the active code.
2479 Print executed "if" and "elsif" statements when debugging is used. This
2480 helps a great deal to understand what a filter does.
2482 Document more things not specified clearly in RFC3028. I had all this
2483 sorted out, when out of a sudden new issues came to my mind. Oops."
2485 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
2488 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
2489 canonical form (as documented). However, after a host name lookup from
2490 the IP address, check_host() was doing a simple string comparison with
2491 addresses acquired from the DNS when checking that the found name did
2492 have the original IP as one of its addresses. Since any found IPv6
2493 addresses are likely to be in abbreviated form, the comparison could
2494 fail. Luckily, there already exists a function for doing the comparison
2495 by converting both addresses to binary, so now that is used instead of
2496 the text comparison.
2498 PH/20 There was another similar case to PH/19, when a complete host name was
2499 given in a host list; looking up its IP address could give an abbreviated
2500 form, whereas the current host's name might or might not be abbreviated.
2501 The same fix has been applied.
2507 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
2508 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
2511 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
2512 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
2514 PH/03 Typo: missing ".o" in src/pcre/Makefile.
2516 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
2517 header line, restrict the check to what is listed in RFCs 2369 and 2929.
2518 Also, for "Auto-Submitted", treat anything other than "no" as
2519 non-personal, in accordance with RFC 3834. (Previously it treated
2520 anything starting "auto-" as non-personal.)
2522 TF/01 The control=submission/name=... option had a problem with syntax
2523 errors if the name included a slash character. The /name= option
2524 now slurps the rest of the string, so it can include any characters
2525 but it must come last in the list of options (after /sender_retain
2528 PH/05 Some modifications to the interface to the fake nameserver for the new
2536 TK/01 Added the "success_on_redirect" address verification option. See
2537 NewStuff for rationale and an example.
2539 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
2541 PH/02 Patch to exigrep to allow it to work on syslog lines.
2543 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
2544 fread() to skip over the body file's header line, because in Cygwin the
2545 header line is locked and is inaccessible.
2547 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
2548 co-exist for some time) to make it clear that it is the Exim ID that is
2549 referenced, not the Message-ID: header line.
2551 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
2552 string_format() function, because snprintf() does not exist on all
2555 PH/06 The use of forbid_filter_existstest now also locks out the use of the
2556 ${stat: expansion item.
2558 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
2559 protocol synchronization error", to keep the pedants happy.
2561 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
2562 well as for IRIX systems, when gcc is being used. See the host.c source
2565 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
2567 PH/10 Named domain lists were not working if used in a queue_smtp_domains
2570 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
2571 transport and to the smtp transport in LMTP mode.
2573 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
2575 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
2576 run a filter in a subprocess. This could lead to confusion in subsequent
2577 lookups in the parent process. There should also be a search_tidyup() at
2578 the end of the subprocess.
2580 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
2581 only if the host matched helo_try_verify_hosts, which caused the
2582 verification to occur when the EHLO/HELO command was issued. The ACL just
2583 tested the remembered result. Now, if a previous verification attempt has
2584 not happened, "verify = helo" does it there and then.
2586 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
2588 TK/03 Fix log output including CR from clamd.
2590 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
2591 exist provoked a memory error which could cause a segfault.
2593 PH/15 Installed PCRE 6.2
2595 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
2597 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
2598 of the problem. Specifically, suggested +O2 rather than +O1 for the
2601 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
2603 PH/20 If a delivery was routed to a non-standard port by means of an SRV
2604 record, the port was not correctly logged when the outgoing_port log
2605 selector was set (it logged the transort's default port).
2607 PH/21 Added support for host-specific ports to manualroute, queryprogram,
2608 fallback_hosts, and "hosts" in the smtp transport.
2610 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
2611 host errors such as "Connection refused".
2613 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
2614 authentication with radiusclient 0.4.9:
2616 - Error returned from rc_read_config was caught wrongly
2617 - Username/password not passed on to radius server due to wrong length.
2619 The presumption is that some radiusclient API changes for 4.51/PH/17
2620 were not taken care of correctly. The code is still untested by me (my
2621 Linux distribution still has 0.3.2 of radiusclient), but it was
2622 contributed by a Radius user.
2624 PH/24 When doing a callout, the value of $domain wasn't set correctly when
2625 expanding the "port" option of the smtp transport.
2627 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
2628 while reading a MIME header. Thanks to Tom Hughes for a patch.
2630 PH/24 Include config.h inside local_scan.h so that configuration settings are
2633 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
2634 that in an ACL for RCPT (for example), you can examine exactly what was
2637 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
2638 commands, but it was not correctly comparing the address with the actual
2639 client host address. Thus, it would show the EHLO address in Received:
2640 header lines when this was not necessary.
2642 PH/27 Added the % operator to ${eval:}.
2644 PH/28 Exim tries to create and chdir to its spool directory when it starts;
2645 it should be ignoring failures (because with -C, for example, it has lost
2646 privilege). It wasn't ignoring creation failures other than "already
2649 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
2652 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
2653 because an input line was too long, either on its own, or by virtue of
2654 too many continuations, the temporary file was not being removed, and the
2655 return code was incorrect.
2657 PH/31 Missing "BOOL" in function definition in filtertest.c.
2659 PH/32 Applied Sieve patches from the maintainer.
2661 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
2663 PH/33 Added "verify = not_blind".
2665 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
2666 Local/Makefile (with some defaults set). These are used in built scripts
2667 such as exicyclog, but they have never been used in the exim_install
2668 script (though there are many overriding facilities there). I have
2669 arranged that the exim_install script now takes note of these two
2672 PH/35 Installed configuration files for Dragonfly.
2674 PH/36 When a locally submitted message by a trusted user did not contain a
2675 From: header, and the sender address was obtained from -f or from an SMTP
2676 MAIL command, and the trusted user did not use -F to supply a sender
2677 name, $originator_name was incorrectly used when constructing a From:
2678 header. Furthermore, $originator_name was used for submission mode
2679 messages from external hosts without From: headers in a similar way,
2680 which is clearly wrong.
2682 PH/37 Added control=suppress_local_fixups.
2684 PH/38 When log_selector = +received_sender was set, and the addition of the
2685 sender made the log line's construction buffer exactly full, or one byte
2686 less than full, an overflow happened when the terminating "\n" was
2689 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
2690 when the result of a list match is failure because a DNS lookup failed.
2692 PH/40 RM_COMMAND is now used in the building process.
2694 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
2695 the "build-* directories that it finds.
2697 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
2698 address in a domain literal was a prefix of an interface address.
2700 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
2701 when verifying a sender address, unless rewrite_headers is false.
2703 PH/44 Wrote a long comment about why errors_to addresses are verified as
2704 recipients, not senders.
2706 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
2707 the ratelimit ACL was added.
2709 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
2711 PH/46 Added extra information about PostgreSQL errors to the error string.
2713 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
2714 avoiding the need to install special zones in a real server. This is
2715 backwards compatible; if it can't find the fake resolver, it drops back.
2716 Thus, both old and new test suites can be run.
2718 TF/02 Added util/ratelimit.pl
2720 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
2721 clock is set back in time.
2723 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
2724 Candler <B.Candler@pobox.com>.
2726 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
2727 OK for addresses that are the result of redirections.
2729 PH/48 A number of further additions for the benefit of the new test suite,
2730 including a fake gethostbyname() that interfaces to the fake DNS resolver
2733 TF/06 The fix for widen_domains has also been applied to qualify_single and
2734 search_parents which are the other dnslookup options that can cause
2737 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
2740 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
2742 TF/07 Exim produced the error message "an SRV record indicated no SMTP
2743 service" if it encountered an MX record with an empty target hostname.
2744 The message is now "an MX or SRV record indicated no SMTP service".
2746 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
2747 if the DNS of the sending site is misconfigured. This is quite a
2748 common situation. This change restores the behaviour of treating a
2749 helo verification defer as a failure.
2751 PH/51 If self=fail was set on a router, the bounce message did not include the
2752 actual error message.
2758 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
2760 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
2761 command itself ended in error, the underlying message about the transport
2762 filter timeout was being overwritten with the pipe command error. Now the
2763 underlying error message should be appended to the second error message.
2765 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
2767 PH/02 Reduce the amount of output that "make" produces by default. Full output
2768 can still be requested.
2770 PH/03 The warning log line about a condition test deferring for a "warn" verb
2771 was being output only once per connection, rather than after each
2772 occurrence (because it was using the same function as for successful
2773 "warn" verbs). This seems wrong, so I have changed it.
2775 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
2776 it should not have, which might have caused a crash in the right
2777 circumstances, but probably never did.
2779 PH/04 Installed a modified version of Tony Finch's patch to make submission
2780 mode fix the return path as well as the Sender: header line, and to
2781 add a /name= option so that you can make the user's friendly name appear
2784 TF/03 Added the control = fakedefer ACL modifier.
2786 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
2787 Mark Lowes for thorough testing.
2789 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
2791 TK/03 Merged latest SRS patch from Miles Wilton.
2793 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
2794 with the definition in sysexits.h (which is #included earlier).
2795 Fortunately, Exim does not actually use EX_OK. The code used to try to
2796 preserve the sysexits.h value, by assumimg that macro definitions were
2797 scanned for macro replacements. I have been disabused of this notion,
2798 so now the code just undefines EX_OK before #including unistd.h.
2800 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
2801 in the smtp transport. When a block could not be written in a single
2802 write() function, the timeout was being re-applied to each part-write.
2803 This seems wrong - if the receiver was accepting one byte at a time it
2804 would take for ever. The timeout is now adjusted when this happens. It
2805 doesn't have to be particularly precise.
2807 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
2808 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
2810 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
2811 without a domain if the check to <postmaster@domain> fails.
2813 SC/01 Eximstats: added -xls and the ability to specify output files
2814 (patch written by Frank Heydlauf).
2816 SC/02 Eximstats: use FileHandles for outputing results.
2818 SC/03 Eximstats: allow any combination of xls, txt, and html output.
2820 SC/04 Eximstats: fixed display of large numbers with -nvr option
2822 SC/05 Eximstats: fixed merging of reports with empty tables.
2824 SC/06 Eximstats: added the -include_original_destination flag
2826 SC/07 Eximstats: removed tabs and trailing whitespace.
2828 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
2830 TK/06 MBOX spool code: Add real "From " MBOX separator line
2831 so the .eml file is really in mbox format (even though
2832 most programs do not really care). Patch from Alex Miller.
2834 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
2835 The latter is generated from $received_to and is only set if the
2836 message has one envelope recipient. SA can use these headers,
2837 obviously out-of-the-box. Patch from Alex Miller.
2839 PH/08 The ${def test on a variable was returning false if the variable's
2840 value was "0", contrary to what the specification has always said!
2841 The result should be true unless the variable is empty.
2843 PH/09 The syntax error of a character other than { following "${if
2844 def:variable_name" (after optional whitespace) was not being diagnosed.
2845 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
2846 accidental colon was present, for example, could give incorrect results.
2848 PH/10 Tidied the code in a number of places where the st_size field of a stat()
2849 result is used (not including appendfile, where other changes are about
2852 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
2853 This involved changing a lot of size variables from int to off_t. It
2854 should work with maildirs and everything.
2856 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
2857 spamd dying while we are connected to it.
2859 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
2862 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
2863 function for BDB 4.3.
2865 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
2867 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
2868 the sources for PCRE have been re-arranged and more files are now
2871 PH/15 The code I had for printing potentially long long variables in PH/11
2872 above was not the best (it lost precision). The length of off_t variables
2873 is now inspected at build time, and an appropriate printing format (%ld
2874 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
2875 to be "long long int" or "long int". This is needed for the internal
2876 formatting function string_vformat().
2878 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
2879 the configuration file to be ":syslog", then the script "guesses" where
2880 the logs files are, rather than using the compiled in default. In our
2881 case the guess is not the same as the compiled default, so the script
2882 suddenly stopped working when I started to use syslog. The patch checks
2883 to see if log_file_path is "". If so, it attempts to read it from exim
2884 with no configuration file to get the compiled in version, before it
2885 falls back to the previous guessing code."
2887 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
2888 implementing BATV in an Exim configuration. See NewStuff for the gory
2891 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
2892 Makefile that are specific to HP-UX.
2894 PH/18 If the "use_postmaster" option was set for a recipient callout together
2895 with the "random" option, the postmaster address was used as the MAIL
2896 FROM address for the random test, but not for the subsequent recipient
2897 test. It is now used for both.
2899 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
2900 patch removes a few documentation additions to RFC 3028, because the
2901 latest draft now contains them. It adds the new en;ascii-case comparator
2902 and a new error check for 8bit text in MIME parts. Comparator and
2903 require names are now matched exactly. I enabled the subaddress
2904 extension, but it is not well tested yet (read: it works for me)."
2906 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
2907 rework some of the code of TK/09 above to avoid the hardwired use of
2908 "%lld" and "long long". Replaced the call to snprintf() with a call to
2911 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
2912 records point to non-existent hosts", "retry timeout exceeded", and
2913 "retry time not reached for any host after a long failure period".
2915 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2916 experimental DomainKeys support:
2918 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2919 (2) On an error such as an illegally used "control", the wrong name for
2920 the control was given.
2922 These problems did NOT occur unless DomainKeys support was compiled.
2924 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2926 PH/24 Added ${if match_ip condition.
2928 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2929 databases so that it will be absolutely obvious if a crash occurs in the
2930 DB library. This is a regular occurrence (often caused by mis-matched
2933 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2934 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2935 picked up on a user's system that detects such things. There doesn't seem
2936 to be a gcc warning option for this - only an attribute that has to be
2937 put on the function's prototype. It seems that in Fedora Core 4 they have
2938 set this on a number of new functions. No doubt there will be more in due
2941 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2942 specify a transport. However, if an address that was verified by such a
2943 router was the subject of a callout, Exim crashed because it tried to
2944 read the rcpt_include_affixes from the non-existent transport. Now it
2945 just assumes that the setting of that option is false. This bug was
2946 introduced by 4.51/PH/31.
2948 PH/28 Changed -d+all to exclude +memory, because that information is very
2949 rarely of interest, but it makes the output a lot bigger. People tend to
2950 do -d+all out of habit.
2952 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2953 code in os-type was giving problems when libc.so lives in lib64, like on
2956 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2957 aren't the modern standard, and it seems that some systems' include files
2958 don't always have them. Exim was already checking for some of the newer
2959 ones like T_AAAA, and defining it itself. I've added checks for all the
2960 record types that Exim uses.
2962 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2963 not automatically generating a new one, as it is supposed to. This
2964 prevented TLS from working. If the file did exist, but contained invalid
2965 data, a new version was generated, as expected. It was only the case of a
2966 non-existent file that was broken.
2968 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2969 with a change in libdomainkeys > 0.64.
2971 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2972 from DNS. If the selector record carries the flag, it now has
2973 precedence over the domain-wide flag.
2975 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2977 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2978 the use of a port name that isn't defined in /etc/services) occurred, the
2979 message was deferred as in a normal delivery, and thus remained on the
2980 spool, instead of being failed because of the mua_wrapper setting. This
2981 is now fixed, and I tidied up some of the mua_wrapper messages at the
2984 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2985 the messages in a hash of arrays rather than using individual hashes.
2986 This is a bit cleaner and results in dramatic memory savings, albeit
2987 at a slight CPU cost.
2989 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2990 as requested by Marc Sherman.
2992 SC/10 Eximstats: added histograms for user specified patterns as requested
2995 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2997 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2998 fopen() in the content-scanning modules that did not already have it.
3004 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3005 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3007 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3009 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3011 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3012 to test Sieve filters that use "vacation".
3014 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3015 that changes the way the GnuTLS parameters are stored in the cache file.
3016 The new format can be generated externally. For backward compatibility,
3017 if the data in the cache doesn't make sense, Exim assumes it has read an
3018 old-format file, and it generates new data and writes a new file. This
3019 means that you can't go back to an older release without removing the
3022 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3023 work if there are any delivery delays because "one_time" forces the
3024 parent to be marked "delivered", so its unseen clone is never tried
3025 again. For this reason, Exim now forbids the simultaneous setting of
3028 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3029 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3030 entry if you want to know the details. The fix, however, neglected to
3031 consider the case when local delivery batching is involved. The test for
3032 "previously delivered" was not happening when checking to see if an
3033 address could be batched with a previous (undelivered) one; under
3034 certain circumstances this could lead to multiple deliveries to the same
3037 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3038 in its include files, and this causes problems building Exim.
3040 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3041 header_syntax) but Exim was just ignoring anything given after a slash.
3042 In particular, this caused confusion with an attempt to use "verify =
3043 reverse_host_lookup/defer_ok". An error is now given when options are
3044 supplied for verify items that do not have them. (Maybe reverse_host_
3045 lookup should have a defer_ok option, but that's a different point.)
3047 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3048 defined by RFC 821) to 2048, because there were problems with some AUTH
3049 commands, and RFC 1869 says the size should be increased for extended
3050 SMTP commands that take arguments.
3052 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3055 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3056 "unknown" error; now it says that the functionality isn't in the binary.
3058 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3059 an address' error message when a string expansion fails (syntax or
3060 whatever). Otherwise the password may appear in the log. Following change
3061 PH/42 below, there is no longer a chance of it appearing in a bounce
3064 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3066 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3067 of its addresses was ever tried. (Bugzilla bug #2.)
3069 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3070 the result incorrectly in the debug output. (It correctly added a newline
3071 to what was transported.)
3073 TF/01 Added $received_time.
3075 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3076 commented out examples of how to interface to a virus scanner and to
3077 SpamAssassin. Also added commented examples of av_scanner and
3078 spamd_address settings.
3080 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3081 and controls are allowed in which ACLs. There were a couple of minor
3082 errors. Some of the entries in the conditions table (which is a table of
3083 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3084 negation of where the condition IS allowed.
3086 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3088 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3089 header file does not have a version number, so I've had to invent a new
3090 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3091 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3092 radiusclient), but it was contributed by a Radius user.
3094 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3095 files or directories, for OpenSSL.
3097 PH/19 When an Exim process that is running as root has to create an Exim log
3098 file, it does so in a subprocess that runs as exim:exim so as to get the
3099 ownership right at creation (otherwise, other Exim processes might see
3100 the file with the wrong ownership). There was no test for failure of this
3101 fork() call, which would lead to the process getting stuck as it waited
3102 for a non-existent subprocess. Forks do occasionally fail when resources
3103 run out. I reviewed all the other calls to fork(); they all seem to check
3106 PH/20 When checking for unexpected SMTP input at connect time (before writing
3107 the banner), Exim was not dealing correctly with a non-positive return
3108 from the read() function. If the client had disconnected by this time,
3109 the result was a log entry for a synchronization error with an empty
3110 string after "input=" when read() returned zero. If read() returned -1
3111 (an event I could not check), uninitialized data bytes were printed.
3112 There were reports of junk text (parts of files, etc) appearing after
3115 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3117 PH/22 Added support for macro redefinition, and (re)definition in between
3118 driver and ACL definitions.
3120 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3121 forgetting to use the resulting value; it was using the unexpanded value.
3123 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3124 hadn't been configured. The fix is from Juergen Kreileder, who
3125 understands it better than I do:
3127 "Here's what I see happening with three configured cyrus_sasl
3128 authenticators configured (plain, login, cram-md5):
3130 On startup auth_cyrus_sasl_init() gets called for each of these.
3131 This means three calls to sasl_listmech() without a specified mech_list.
3132 => SASL tests which mechs of all available mechs actually work
3133 => three warnings about OTP not working
3134 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3136 With the patch, sasl_listmech() also gets called three times. But now
3137 SASL's mech_list option is set to the server_mech specified in the the
3138 authenticator. Or in other words, the answer from sasl_listmech()
3139 gets limited to just the mech you're testing for (which is different
3141 => the return list contains just 'plain' or 'login', 'cram-md5' or
3142 nothing depending on the value of ob->server_mech.
3144 I've just tested the patch: Authentication still works fine,
3145 unavailable mechs specified in the exim configuration are still
3146 caught, and the auth.log warnings about OTP are gone."
3148 PH/25 When debugging is enabled, the contents of the command line are added
3149 to the debugging output, even when log_selector=+arguments is not
3152 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3153 answer is "GNU", and only if the return is "GNU/something" is the answer
3156 PH/27 $acl_verify_message is now set immediately after the failure of a
3157 verification in an ACL, and so is available in subsequent modifiers. In
3158 particular, the message can be preserved by coding like this:
3160 warn !verify = sender
3161 set acl_m0 = $acl_verify_message
3163 Previously, $acl_verify_message was set only while expanding "message"
3164 and "log_message" when a very denied access.
3166 PH/28 Modified OS/os.c-Linux with
3168 -#ifndef OS_LOAD_AVERAGE
3169 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3171 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3172 nomenclature these days.)
3174 PH/29 Installed patch from the Sieve maintainer that adds the options
3175 sieve_useraddress and sieve_subaddress to the redirect router.
3177 PH/30 In these circumstances:
3178 . Two addresses routed to the same list of hosts;
3179 . First host does not offer TLS;
3180 . First host accepts first address;
3181 . First host gives temporary error to second address;
3182 . Second host offers TLS and a TLS session is established;
3183 . Second host accepts second address.
3184 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3185 and peerdn, if requested) that were in fact used only for the second
3188 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3189 attention to any local part prefix or suffix that was matched by the
3190 router that accepted the address. It now behaves in the same way as it
3191 does for delivery: the affixes are removed from the local part unless
3192 rcpt_include_affixes is set on the transport.
3194 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3195 timeout during the DATA phase of an incoming message.
3197 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3198 applied a patch sanctioned by the Sieve maintainer.
3200 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3201 the uid or gid is negative. A case of a negative gid caused this to be
3202 noticed. The fix allows for either to be negative.
3204 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3205 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3208 PH/36 The change PH/12 above was broken. Fixed it.
3210 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3211 the grounds that routing the same address twice would always produce the
3212 same answer. This might have been true once, but it is certainly no
3213 longer true now. Routing a child address may depend on the previous
3214 routing that produced that child. Some complicated redirection strategies
3215 went wrong when messages had multiple recipients, and made Exim's
3216 behaviour dependent on the order in which the addresses were given.
3218 I have moved the duplicate checking until after the routing is complete.
3219 Exim scans the addresses that are assigned to local and remote
3220 transports, and removes any duplicates. This means that more work will be
3221 done, as duplicates will always all be routed, but duplicates are
3222 presumably rare, so I don't expect this is of any significance.
3224 For deliveries to pipes, files, and autoreplies, the duplicate checking
3225 still happens during the routing process, since they are not going to be
3228 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3229 It corrects a timeout issue with spamd. This is Ian's comment: "The
3230 background is that sometimes spamd either never reads data from a
3231 connection it has accepted, or it never writes response data. The exiscan
3232 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3233 blindly assumes that writes won't block so it may never time out."
3235 PH/39 Allow G after quota size as well as K and M.
3237 PH/40 The value set for $authenticated_id in an authenticator may not contain
3238 binary zeroes or newlines because the value is written to log lines and
3239 to spool files. There was no check on this. Now the value is run through
3240 the string_printing() function so that such characters are converted to
3241 printable escape sequences.
3243 PH/41 $message_linecount is a new variable that contains the total number of
3244 lines in the message. Compare $body_linecount, which is the count for the
3247 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3248 bounce and delay warning messages, except in certain special cases, which
3251 (a) An SMTP error message from a remote host;
3252 (b) A message specified in a :fail: redirection;
3253 (c) A message specified in a "fail" command in a system filter;
3254 (d) A message specified in a FAIL return from the queryprogram router;
3255 (e) A message specified by the cannot_route_message router option.
3257 In these cases only, Exim does include the error details in bounce and
3258 warning messages. There are also a few cases where bland messages such
3259 as "unrouteable address" or "local delivery error" are given.
3261 PH/43 $value is now also set for the "else" part of a ${run expansion.
3263 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3264 being worked on, but at least Exim now implements the latest version to
3267 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3268 process to complete was treated as a delivery failure, a timeout while
3269 writing the message to the pipe was logged, but erroneously treated as a
3270 successful delivery. Such timeouts include transport filter timeouts. For
3271 consistency with the overall process timeout, these timeouts are now
3272 treated as errors, giving rise to delivery failures by default. However,
3273 there is now a new Boolean option for the pipe transport called
3274 timeout_defer, which, if set TRUE, converts the failures into defers for
3275 both kinds of timeout. A transport filter timeout is now identified in
3278 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3279 systems where "make" and "gmake" are different, calling "gmake" at top
3280 level broke things. I've arranged for the value of $(MAKE) to be passed
3281 from the Makefile to this script so that it can call the same version of
3285 A note about Exim versions 4.44 and 4.50
3286 ----------------------------------------
3288 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3289 changes of various kinds. As a consequence, a big documentation update was
3290 needed. This delayed the release for rather longer than seemed good, especially
3291 in the light of a couple of (minor) security issues. Therefore, the changes
3292 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3293 release. So 4.44 and 4.50 are in effect two different branches that both start
3296 I have left the 4.50 change log unchanged; it contains all the changes since
3297 4.43. The change log for 4.44 is below; many of its items are identical to
3298 those for 4.50. This seems to be the most sensible way to preserve the
3299 historical information.
3305 1. Minor wording change to the doc/README.SIEVE file.
3307 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3308 computation of the current number of files was incorrect.
3310 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3311 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3314 4. Give more explanation in the error message when the command for a transport
3315 filter fails to execute.
3317 5. There are several places where Exim runs a non-Exim command in a
3318 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3319 was being done only for the command run by the queryprogram router. It is
3320 now done for all such subprocesses. The other cases are: ${run, transport
3321 filters, and the commands run by the lmtp and pipe transports.
3323 6. Added CONFIGURE_GROUP build-time option.
3325 7. Some older OS have a limit of 256 on the maximum number of file
3326 descriptors. Exim was using setrlimit() to set 1000 as a large value
3327 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3328 systems. I've change it so that if it can't get 1000, it tries for 256.
3330 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3331 was an oversight, and furthermore, ever since the addition of extra
3332 controls (e.g. 4.43/32), the checks on when to allow different forms of
3333 "control" were broken. There should now be diagnostics for all cases when a
3334 control that does not make sense is encountered.
3336 9. Added the /retain_sender option to "control=submission".
3338 10. $recipients is now available in the predata ACL (oversight).
3340 11. Tidy the search cache before the fork to do a delivery from a message
3341 received from the command line. Otherwise the child will trigger a lookup
3342 failure and thereby defer the delivery if it tries to use (for example) a
3343 cached ldap connection that the parent has called unbind on.
3345 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3346 of $address_data from the recipient verification was clobbered by the
3347 sender verification.
3349 13. The value of address_data from a sender verification is now available in
3350 $sender_address_data in subsequent conditions in the ACL statement.
3352 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3354 15. Added a new option "connect=<time>" to callout options, to set a different
3357 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3358 was its contents. (It was OK if the option was not defined at all.)
3360 17. A "Completed" log line is now written for messages that are removed from
3361 the spool by the -Mrm option.
3363 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3364 information about exactly what failed.
3366 19. Added -dd to debug only the daemon process.
3368 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3369 handles timeouts, both on the server side and network timeouts. Renamed the
3370 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3372 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3373 It is now set to "smtps".
3375 22. $host_address is now set to the target address during the checking of
3376 ignore_target_hosts.
3378 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3379 being passed; this would have caused $sender_host_name to have been used if
3380 matching the list had actually called for a host name (not very likely,
3381 since this list is usually IP addresses). A host name is now passed as
3384 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3385 code that specifies a non-restarting handler (typically sigaction() in
3386 modern systems) in an attempt to fix a rare and obscure crash bug.
3388 25. Narrowed the window for a race in the daemon that could cause it to ignore
3389 SIGCHLD signals. This is not a major problem, because they are used only to
3390 wake it up if nothing else does.
3392 26. A malformed maildirsize file could cause Exim to calculate negative values
3393 for the mailbox size or file count. Odd effects could occur as a result.
3394 The maildirsize information is now recalculated if the size or filecount
3397 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3398 support for a long time. Removed HAVE_SYS_VFS_H.
3400 28. Installed the latest version of exipick from John Jetmore.
3402 29. In an address list, if the pattern was not a regular expression, an empty
3403 subject address (from a bounce message) matched only if the pattern was an
3404 empty string. Non-empty patterns were not even tested. This was the wrong
3405 because it is perfectly reasonable to use an empty address as part of a
3406 database query. An empty address is now tested by patterns that are
3407 lookups. However, all the other forms of pattern expect the subject to
3408 contain a local part and a domain, and therefore, for them, an empty
3409 address still always fails if the pattern is not itself empty.
3411 30. Exim went into a mad DNS loop when attempting to do a callout where the
3412 host was specified on an smtp transport, and looking it up yielded more
3413 than one IP address.
3415 31. Re-factored the code for checking spool and log partition space into a
3416 function that finds that data and another that does the check. The former
3417 is then used to implement four new variables: $spool_space, $log_space,
3418 $spool_inodes, and $log_inodes.
3420 32. The RFC2047 encoding function was originally intended for short strings
3421 such as real names; it was not keeping to the 75-character limit for
3422 encoded words that the RFC imposes. It now respects the limit, and
3423 generates multiple encoded words if necessary. To be on the safe side, I
3424 have increased the buffer size for the ${rfc2047: expansion operator from
3427 33. It is now permitted to omit both strings after an "if" condition; if the
3428 condition is true, the result is "true". As before, when the second string
3429 is omitted, a false condition yields an empty string. This makes it less
3430 cumbersome to write custom ACL and router conditions.
3432 34. Failure to deliver a bounce message always caused it to be frozen, even if
3433 there was an errors_to setting on the router. The errors_to setting is now
3436 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3437 canonical form (fully expanded) before being placed in
3438 $sender_host_address.
3440 36. The table in the code that translates DNS record types into text (T_A to
3441 "A" for instance) was missing entries for NS and CNAME. It is just possible
3442 that this could have caused confusion if both these types were looked up
3443 for the same domain, because the text type is used as part of Exim's
3444 per-process caching. But the chance of anyone hitting this buglet seems
3447 37. The dnsdb lookup has been extended in a number of ways.
3449 (1) There is a new type, "zns", which walks up the domain tree until it
3450 finds some nameserver records. It should be used with care.
3452 (2) There is a new type, "mxh", which is like "mx" except that it returns
3453 just the host names, not the priorities.
3455 (3) It is now possible to give a list of domains (or IP addresses) to be
3456 looked up. The behaviour when one of the lookups defers can be
3457 controlled by a keyword.
3459 (4) It is now possible to specify the separator character for use when
3460 multiple records are returned.
3462 38. The dnslists ACL condition has been extended: it is now possible to supply
3463 a list of IP addresses and/or domains to be looked up in a particular DNS
3466 39. Added log_selector=+queue_time_overall.
3468 40. When running the queue in the test harness, wait just a tad after forking a
3469 delivery process, to get repeatability of debugging output.
3471 41. Include certificate and key file names in error message when GnuTLS fails
3472 to set them up, because the GnuTLS error message doesn't include the name
3473 of the failing file when there is a problem reading it.
3475 42. Allow both -bf and -bF in the same test run.
3477 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
3479 44. The "Exiscan patch" is now merged into the mainline Exim source.
3481 45. Sometimes the final signoff response after QUIT could fail to get
3482 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3483 < 0 before doing a fflush(). This bug looks as though it goes back to the
3484 introduction of TLS in release 3.20, but "sometimes" must have been rare
3485 because the tests only now provoked it.
3487 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
3488 (this can affect the format of dates).
3490 47. exim_tidydb, when checking for the continued existence of a message for
3491 which it has found a message-specific retry record, was not finding
3492 messages that were in split spool directories. Consequently, it was
3493 deleting retry records that should have stayed in existence.
3495 48. Steve fixed some bugs in eximstats.
3497 49. The SPA authentication driver was not abandoning authentication and moving
3498 on to the next authenticator when an expansion was forced to fail,
3499 contradicting the general specification for all authenticators. Instead it
3500 was generating a temporary error. It now behaves as specified.
3502 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
3503 (the order specifies the preference for clients). The order is now AES256,
3504 AES128, 3DES, ARCFOUR128.
3506 51. Small patch to Sieve code - explicitly set From: when generating an
3509 52. Exim crashed if a remote delivery caused a very long error message to be
3510 recorded - for instance if somebody sent an entire SpamAssassin report back
3511 as a large number of 550 error lines. This bug was coincidentally fixed by
3512 increasing the size of one of Exim's internal buffers (big_buffer) that
3513 happened as part of the Exiscan merge. However, to be on the safe side, I
3514 have made the code more robust (and fixed the comments that describe what
3517 53. Now that there can be additional text after "Completed" in log lines (if
3518 the queue_time_overall log selector is set), a one-byte patch to exigrep
3519 was needed to allow it to recognize "Completed" as not the last thing in
3522 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
3523 patch that reportedly fixes this has been added. I am not expert enough to
3524 create a test for it. This is what the patch creator wrote:
3526 "I found a little strange behaviour of ldap code when working with
3527 Windows 2003 AD Domain, where users was placed in more than one
3528 Organization Units. When I tried to give exim partial DN, the exit code
3529 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
3530 But simultaneously result of request was absolutely normal ldap result,
3531 so I produce this patch..."
3533 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
3534 REFERENCE, so I have modified the code to exclude the patch when that macro
3537 55. Some experimental protocols are using DNS PTR records for new purposes. The
3538 keys for these records are domain names, not reversed IP addresses. The
3539 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3540 leaves it alone. Component reversal etc. now happens only for IP addresses.
3543 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
3545 57. Double the size of the debug message buffer (to 2048) so that more of very
3546 long debug lines gets shown.
3548 58. The exicyclog utility now does better if the number of log files to keep
3549 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
3551 59. Two changes related to the smtp_active_hostname option:
3553 (1) $smtp_active_hostname is now available as a variable.
3554 (2) The default for smtp_banner uses $smtp_active_hostname instead
3555 of $primary_hostname.
3557 60. The host_aton() function is supposed to be passed a string that is known
3558 to be a valid IP address. However, in the case of IPv6 addresses, it was
3559 not checking this. This is a hostage to fortune. Exim now panics and dies
3560 if the condition is not met. A case was found where this could be provoked
3561 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3562 components; fortuitously, this particular loophole had already been fixed
3563 by change 4.50/55 above.
3565 If there are any other similar loopholes, the new check in host_aton()
3566 itself should stop them being exploited. The report I received stated that
3567 data on the command line could provoke the exploit when Exim was running as
3568 exim, but did not say which command line option was involved. All I could
3569 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3570 running as the user.
3573 61. There was a buffer overflow vulnerability in the SPA authentication code
3574 (which came originally from the Samba project). I have added a test to the
3575 spa_base64_to_bits() function which I hope fixes it.
3578 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
3579 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
3581 63. The daemon start-up calls getloadavg() while still root for those OS that
3582 need the first call to be done as root, but it missed one case: when
3583 deliver_queue_load_max is set with deliver_drop_privilege. This is
3584 necessary for the benefit of the queue runner, because there is no re-exec
3585 when deliver_drop_privilege is set.
3587 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
3588 This has been fixed.
3590 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3591 was in use, was not putting the data itself into the right store pool;
3592 consequently, it could be overwritten for a subsequent message in the same
3593 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3596 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
3598 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
3599 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
3600 the function were treating the return as a boolean value, which happened to
3601 work because 0=false and not-0=true, but is not correct code.
3603 68. The host_aton() function was not handling scoped IPv6 addresses (those
3604 with, for example, "%eth0" on the end) correctly.
3606 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
3607 negated items (that is, ~something) in unsigned ints. Some compilers
3608 apparently mutter when there is no cast.
3610 70. If an address verification called from an ACL failed, and did not produce a
3611 user-specific message (i.e. there was only a "system" message), nothing was
3612 put in $acl_verify_message. In this situation, it now puts the system
3615 71. Change 4.23/11 added synchronization checking at the start of an SMTP
3616 session; change 4.31/43 added the unwanted input to the log line - except
3617 that it did not do this in the start of session case. It now does.
3619 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
3620 This is wrong and can cause the other end to generate a synchronization
3621 error if it is another Exim or anything else that does the synchronization
3622 check. A QUIT command is no longer sent after a timeout.
3624 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
3625 during host lookups.
3627 74. The defer_ok option of callout verification was not working if it was used
3628 when verifying addresses in header lines, that is, for this case:
3630 verify = header_sender/callout=defer_ok
3632 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
3633 those file descriptors could be used for SMTP connections. If anything
3634 wrote to stderr (the example that came up was "warn" in embedded Perl), it
3635 could be sent to the SMTP client, causing chaos. The daemon now opens
3636 stdin, stdout, and stderr to /dev/null when it puts itself into the
3639 76. Arrange for output from Perl's "warn" command to be written to Exim's main
3640 log by default. The user can override this with suitable Perl magic.
3642 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
3643 the log message when discard triggers, was not working for the DATA ACL or
3644 for the non-SMTP ACL.
3646 78. Error message wording change in sieve.c.
3648 79. If smtp_accept_max_per_host was set, the number of connections could be
3649 restricted to fewer than expected, because the daemon was trying to set up
3650 a new connection before checking whether the processes handling previous
3651 connections had finished. The check for completed processes is now done
3652 earlier. On busy systems, this bug wouldn't be noticed because something
3653 else would have woken the daemon, and it would have reaped the completed
3656 80. If a message was submitted locally by a user whose login name contained one
3657 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
3658 It caused a spool format error. I have fixed the spool reading code. A
3659 related problem was that the "from" clause in the Received: line became
3660 illegal because of the space(s). It is now covered by ${quote_local_part.
3662 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
3665 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
3667 83. Patch from Sieve maintainer to fix unterminated string problem in
3668 "vacation" handling.
3670 84. Some minor changes to the Linux configuration files to help with other
3671 OS variants using glibc.
3673 85. One more patch for Sieve to update vacation handling to latest spec.
3676 ----------------------------------------------------
3677 See the note above about the 4.44 and 4.50 releases.
3678 ----------------------------------------------------
3684 1. Change 4.43/35 introduced a bug that caused file counts to be
3685 incorrectly computed when quota_filecount was set in an appendfile
3688 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3689 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3692 3. Give more explanation in the error message when the command for a transport
3693 filter fails to execute.
3695 4. There are several places where Exim runs a non-Exim command in a
3696 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3697 was being done only for the command run by the queryprogram router. It is
3698 now done for all such subprocesses. The other cases are: ${run, transport
3699 filters, and the commands run by the lmtp and pipe transports.
3701 5. Some older OS have a limit of 256 on the maximum number of file
3702 descriptors. Exim was using setrlimit() to set 1000 as a large value
3703 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3704 systems. I've change it so that if it can't get 1000, it tries for 256.
3706 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3707 was an oversight, and furthermore, ever since the addition of extra
3708 controls (e.g. 4.43/32), the checks on when to allow different forms of
3709 "control" were broken. There should now be diagnostics for all cases when a
3710 control that does not make sense is encountered.
3712 7. $recipients is now available in the predata ACL (oversight).
3714 8. Tidy the search cache before the fork to do a delivery from a message
3715 received from the command line. Otherwise the child will trigger a lookup
3716 failure and thereby defer the delivery if it tries to use (for example) a
3717 cached ldap connection that the parent has called unbind on.
3719 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3720 of $address_data from the recipient verification was clobbered by the
3721 sender verification.
3723 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3724 was its contents. (It was OK if the option was not defined at all.)
3726 11. A "Completed" log line is now written for messages that are removed from
3727 the spool by the -Mrm option.
3729 12. $host_address is now set to the target address during the checking of
3730 ignore_target_hosts.
3732 13. When checking ignore_target_hosts for an ipliteral router, no host name was
3733 being passed; this would have caused $sender_host_name to have been used if
3734 matching the list had actually called for a host name (not very likely,
3735 since this list is usually IP addresses). A host name is now passed as
3738 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3739 code that specifies a non-restarting handler (typically sigaction() in
3740 modern systems) in an attempt to fix a rare and obscure crash bug.
3742 15. Narrowed the window for a race in the daemon that could cause it to ignore
3743 SIGCHLD signals. This is not a major problem, because they are used only to
3744 wake it up if nothing else does.
3746 16. A malformed maildirsize file could cause Exim to calculate negative values
3747 for the mailbox size or file count. Odd effects could occur as a result.
3748 The maildirsize information is now recalculated if the size or filecount
3751 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3752 support for a long time. Removed HAVE_SYS_VFS_H.
3754 18. Updated exipick to current release from John Jetmore.
3756 19. Allow an empty sender to be matched against a lookup in an address list.
3757 Previously the only cases considered were a regular expression, or an
3760 20. Exim went into a mad DNS lookup loop when doing a callout where the
3761 host was specified on the transport, if the DNS lookup yielded more than
3764 21. The RFC2047 encoding function was originally intended for short strings
3765 such as real names; it was not keeping to the 75-character limit for
3766 encoded words that the RFC imposes. It now respects the limit, and
3767 generates multiple encoded words if necessary. To be on the safe side, I
3768 have increased the buffer size for the ${rfc2047: expansion operator from
3771 22. Failure to deliver a bounce message always caused it to be frozen, even if
3772 there was an errors_to setting on the router. The errors_to setting is now
3775 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3776 canonical form (fully expanded) before being placed in
3777 $sender_host_address.
3779 24. Updated eximstats to version 1.33
3781 25. Include certificate and key file names in error message when GnuTLS fails
3782 to set them up, because the GnuTLS error message doesn't include the name
3783 of the failing file when there is a problem reading it.
3785 26. Expand error message when OpenSSL has problems setting up cert/key files.
3788 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
3789 (this can affect the format of dates).
3791 28. exim_tidydb, when checking for the continued existence of a message for
3792 which it has found a message-specific retry record, was not finding
3793 messages that were in split spool directories. Consequently, it was
3794 deleting retry records that should have stayed in existence.
3796 29. eximstats updated to version 1.35
3797 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
3798 1.35 - bugfix such that pie charts by volume are generated correctly
3800 30. The SPA authentication driver was not abandoning authentication and moving
3801 on to the next authenticator when an expansion was forced to fail,
3802 contradicting the general specification for all authenticators. Instead it
3803 was generating a temporary error. It now behaves as specified.
3805 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
3806 (the order specifies the preference for clients). The order is now AES256,
3807 AES128, 3DES, ARCFOUR128.
3809 31. Small patch to Sieve code - explicitly set From: when generating an
3812 32. Exim crashed if a remote delivery caused a very long error message to be
3813 recorded - for instance if somebody sent an entire SpamAssassin report back
3814 as a large number of 550 error lines. This bug was coincidentally fixed by
3815 increasing the size of one of Exim's internal buffers (big_buffer) that
3816 happened as part of the Exiscan merge. However, to be on the safe side, I
3817 have made the code more robust (and fixed the comments that describe what
3820 33. Some experimental protocols are using DNS PTR records for new purposes. The
3821 keys for these records are domain names, not reversed IP addresses. The
3822 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3823 leaves it alone. Component reversal etc. now happens only for IP addresses.
3826 34. The host_aton() function is supposed to be passed a string that is known
3827 to be a valid IP address. However, in the case of IPv6 addresses, it was
3828 not checking this. This is a hostage to fortune. Exim now panics and dies
3829 if the condition is not met. A case was found where this could be provoked
3830 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3831 components; fortuitously, this particular loophole had already been fixed
3832 by change 4.50/55 or 4.44/33 above.
3834 If there are any other similar loopholes, the new check in host_aton()
3835 itself should stop them being exploited. The report I received stated that
3836 data on the command line could provoke the exploit when Exim was running as
3837 exim, but did not say which command line option was involved. All I could
3838 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3839 running as the user.
3842 35. There was a buffer overflow vulnerability in the SPA authentication code
3843 (which came originally from the Samba project). I have added a test to the
3844 spa_base64_to_bits() function which I hope fixes it.
3847 36. The daemon start-up calls getloadavg() while still root for those OS that
3848 need the first call to be done as root, but it missed one case: when
3849 deliver_queue_load_max is set with deliver_drop_privilege. This is
3850 necessary for the benefit of the queue runner, because there is no re-exec
3851 when deliver_drop_privilege is set.
3853 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3854 was in use, was not putting the data itself into the right store pool;
3855 consequently, it could be overwritten for a subsequent message in the same
3856 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3859 38. Sometimes the final signoff response after QUIT could fail to get
3860 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3861 < 0 before doing a fflush(). This bug looks as though it goes back to the
3862 introduction of TLS in release 3.20, but "sometimes" must have been rare
3863 because the tests only now provoked it.
3869 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
3870 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
3871 Now it returns an integer. A number of calls were still expecting a T/F
3872 return. Fortuitously, in all cases, the tests worked in OK situations,
3873 which is the norm. However, things would have gone wrong on any write
3874 failures on the smtp file descriptor. This function is used when sending
3875 messages over SMTP and also when doing verify callouts.
3877 2. When Exim is called to do synchronous delivery of a locally submitted
3878 message (the -odf or -odi options), it no longer closes stderr before doing
3881 3. Implemented the mua_wrapper option.
3883 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
3885 5. Implemented the functions header_remove(), header_testname(),
3886 header_add_at_position(), and receive_remove_recipient(), and exported them
3889 6. If an ACL "warn" statement specified the addition of headers, Exim already
3890 inserted X-ACL-Warn: at the start if there was no header name. However, it
3891 was not making this test for the second and subsequent header lines if
3892 there were newlines in the string. This meant that an invalid header could
3893 be inserted if Exim was badly configured.
3895 7. Allow an ACL "warn" statement to add header lines at the start or after all
3896 the Received: headers, as well as at the end.
3898 8. Added the rcpt_4xx retry error code.
3900 9. Added postmaster_mailfrom=xxx to callout verification option.
3902 10. Added mailfrom=xxxx to the callout verification option, for verify=
3905 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
3906 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
3908 12. Inserted some casts to stop certain compilers warning when using pointer
3909 differences as field lengths or precisions in printf-type calls (mostly
3910 affecting debugging statements).
3912 13. Added optional readline() support for -be (dynamically loaded).
3914 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
3915 same clock tick as a message's arrival, so that its received time was the
3916 same as the "first fail" time on the retry record, and that message
3917 remained on the queue past the ultimate address timeout, every queue runner
3918 would try a delivery (because it was past the ultimate address timeout) but
3919 after another failure, the ultimate address timeout, which should have then
3920 bounced the address, did not kick in. This was a "< instead of <=" error;
3921 in most cases the first failure would have been in the next clock tick
3922 after the received time, and all would be well.
3924 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
3925 being recognized when the domain list was tested by the match_domain
3926 condition in an expansion string.
3928 16. Added the ${str2b64: operator.
3930 17. Exim was always calling setrlimit() to set a large limit for the number of
3931 processes, without checking whether the existing limit was already
3932 adequate. (It did check for the limit on file descriptors.) Furthermore,
3933 errors from getrlimit() and setrlimit() were being ignored. Now they are
3934 logged to the main and panic logs, but Exim does carry on, to try to do its
3935 job under whatever limits there are.
3937 18. Imported PCRE 5.0.
3939 19. Trivial typo in log message " temporarily refused connection" (the leading
3942 20. If the log selector return_path_on_delivery was set and an address was
3943 redirected to /dev/null, the delivery process crashed because it assumed
3944 that a return path would always be set for a "successful" delivery. In this
3945 case, the whole delivery is bypassed as an optimization, and therefore no
3948 21. Internal re-arrangement: the function for sending a challenge and reading
3949 a response while authentication was assuming a zero-terminated challenge
3950 string. It's now changed to take a pointer and a length, to allow for
3951 binary data in such strings.
3953 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3955 23. Exim was not respecting finduser_retries when seeking the login of the
3956 uid under which it was called; it was always trying 10 times. (The default
3957 setting of finduser_retries is zero.) Also, it was sleeping after the final
3958 failure, which is pointless.
3960 24. Implemented tls_on_connect_ports.
3962 25. Implemented acl_smtp_predata.
3964 26. If the domain in control=submission is set empty, Exim assumes that the
3965 authenticated id is a complete email address when it generates From: or
3966 Sender: header lines.
3968 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3969 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3970 chown and chgrp in /bin and hostname in /usr/bin.
3972 28. Exim was keeping the "process log" file open after each use, just as it
3973 does for the main log. This opens the possibility of it remaining open for
3974 long periods when the USR1 signal hits a daemon. Occasional processlog
3975 errors were reported, that could have been caused by this. Anyway, it seems
3976 much more sensible not to leave this file open at all, so that is what now
3979 29. The long-running daemon process does not normally write to the log once it
3980 has entered its main loop, and it closes the log before doing so. This is
3981 so that log files can straightforwardly be renamed and moved. However,
3982 there are a couple of unusual error situations where the daemon does write
3983 log entries, and I had neglected to close the log afterwards.
3985 30. The text of an SMTP error response that was received during a remote
3986 delivery was being truncated at 512 bytes. This is too short for some of
3987 the long messages that one sometimes sees. I've increased the limit to
3990 31. It is now possible to make retry rules that apply only when a message has a
3991 specific sender, in particular, an empty sender.
3993 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3994 it possible to be selective about when SMTP synchronization is enforced.
3996 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3998 32. Implemented hosts_connection_nolog.
4000 33. Added an ACL for QUIT.
4002 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4005 35. Added mailbox_size and mailbox_filecount to appendfile.
4007 36. Added control = no_multiline_responses to ACLs.
4009 37. There was a bug in the logic of the code that waits for the clock to tick
4010 in the case where the clock went backwards by a substantial amount such
4011 that the microsecond fraction of "now" was more than the microsecond
4012 fraction of "then" (but the whole seconds number was less).
4014 38. Added support for the libradius Radius client library this is found on
4015 FreeBSD (previously only the radiusclient library was supported).
4021 1. When certain lookups returned multiple values in the form name=value, the
4022 quoting of the values was not always being done properly. Specifically:
4023 (a) If the value started with a double quote, but contained no whitespace,
4025 (b) If the value contained whitespace other than a space character (i.e.
4026 tabs or newlines or carriage returns) it was not quoted.
4027 This fix has been applied to the mysql and pgsql lookups by writing a
4028 separate quoting function and calling it from the lookup code. The fix
4029 should probably also be applied to nisplus, ibase and oracle lookups, but
4030 since I cannot test any of those, I have not disturbed their existing code.
4032 2. A hit in the callout cache for a specific address caused a log line with no
4033 reason for rejecting RCPT. Now it says "Previous (cached) callout
4034 verification failure".
4036 3. There was an off-by-one bug in the queryprogram router. An over-long
4037 return line was truncated at 256 instead of 255 characters, thereby
4038 overflowing its buffer with the terminating zero. As well as fixing this, I
4039 have increased the buffer size to 1024 (and made a note to document this).
4041 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4042 when Exim is waiting for an SMTP response from a remote server, Exim
4043 restarts its select() call on the socket, thereby resetting its timeout.
4044 This is not a problem when such interrupts are rare. Somebody set up a cron
4045 job to run exiwhat every 2 minutes, which is less than the normal select()
4046 timeout (5 or 10 minutes). This meant that the select() timeout never
4047 kicked in because it was always reset. I have fixed this by comparing the
4048 time when an interrupt arrives with the time at the start of the first call
4049 to select(). If more time than the timeout has elapsed, the interrupt is
4050 treated as a timeout.
4052 5. Some internal re-factoring in preparation for the addition of Sieve
4053 extensions (by MH). In particular, the "personal" test is moved to a
4054 separate function, and given an option for scanning Cc: and Bcc: (which is
4055 not set for Exim filters).
4057 6. When Exim created an email address using the login of the caller as the
4058 local part (e.g. when creating a From: or Sender: header line), it was not
4059 quoting the local part when it contained special characters such as @.
4061 7. Installed new OpenBSD configuration files.
4063 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4064 try to make them clearer.
4066 9. Callout options, other than the timeout value, were being ignored when
4067 verifying sender addresses in header lines. For example, when using
4069 verify = header_sender/callout=no_cache
4071 the cache was (incorrectly) being used.
4073 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4074 only the Cygwin environment.
4076 11. When return_path_on_delivery was set as a log selector, if different remote
4077 addresses in the same message used different return paths and parallel
4078 remote delivery occurred, the wrong values would sometimes be logged.
4079 (Whenever a remote delivery process finished, the return path value from
4080 the most recently started remote delivery process was logged.)
4082 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4083 lines when AUTH and/or TLS are in use. This is the "received protocol"
4084 field. Exim used to use "asmtp" for authenticated SMTP, without any
4085 indication (in the protocol name) for TLS use. Now it follows the RFC and
4086 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4087 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4088 names appear in log lines as well as in Received: header lines.
4090 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4091 extensions, and comparison tests, and to fix some bugs.
4093 14. Changes to the "personal" filter test:
4095 (1) The test was buggy in that it was just doing the equivalent of
4096 "contains" tests on header lines. For example, if a user's address was
4097 anne@some.where, the "personal" test would incorrectly be true for
4099 To: susanne@some.where
4101 This test is now done by extracting each address from the header in turn,
4102 and checking the entire address. Other tests that are part of "personal"
4103 are now done using regular expressions (for example, to check local parts
4104 of addresses in From: header lines).
4106 (2) The list of non-personal local parts in From: addresses has been
4107 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4108 taken from the Sieve specification recommendations.
4110 (3) If the message contains any header line starting with "List-" it is
4111 treated as non-personal.
4113 (4) The test for "circular" in the Subject: header line has been removed
4114 because it now seems ill-conceived.
4116 15. Minor typos in src/EDITME comments corrected.
4118 16. Installed latest exipick from John Jetmore.
4120 17. If headers_add on a router specified a text string that was too long for
4121 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4122 of string_sprintf() is now avoided.
4124 18. $message_body_size was not set (it was always zero) when running the DATA
4125 ACL and the local_scan() function.
4127 19. For the "mail" command in an Exim filter, no default was being set for
4128 the once_repeat time, causing a random time value to be used if "once" was
4129 specified. (If the value happened to be <= 0, no repeat happened.) The
4130 default is now 0s, meaning "never repeat". The "vacation" command was OK
4131 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4132 (I found it when inspecting the code).
4134 20. There is now an overall timeout for performing a callout verification. It
4135 defaults to 4 times the callout timeout, which applies to individual SMTP
4136 commands during the callout. The overall timeout applies when there is more
4137 than one host that can be tried. The timeout is checked before trying the
4138 next host. This prevents very long delays if there are a large number of
4139 hosts and all are timing out (e.g. when the network connections are timing
4140 out). The value of the overall timeout can be changed by specifying an
4141 additional sub-option for "callout", called "maxwait". For example:
4143 verify = sender/callout=5s,maxwait=20s
4145 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4146 to the end before writing, but this should make it even safer).
4148 22. Exim was forgetting that it had advertised PIPELINING for the second and
4149 subsequent messages on an SMTP connection. It was also not resetting its
4150 memory on STARTTLS and an internal HELO.
4152 23. When Exim logs an SMTP synchronization error within a session, it now
4153 records whether PIPELINING has been advertised or not.
4155 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4156 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4157 rather than long int.
4159 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4161 26. Added the never_mail option to autoreply.
4167 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4168 crash if the getsockname() call failed; this can happen if a connection is
4169 closed very soon after it is established. The problem was simply in the
4170 order in which certain operations were done, causing Exim to try to write
4171 to the SMTP stream before it had set up the file descriptor. The bug has
4172 been fixed by making things happen in the correct order.
4178 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4179 before the connection was closed, thus losing the rejection response.
4181 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4182 some early Solaris releases, but causes trouble in current releases where
4183 socklen_t is defined.
4185 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4188 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4191 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4192 that releasing the top of it at the end releases what was used for sub-
4193 expansions (unless the block got too big). However, discard this block if
4194 the first thing is a variable or header, so that we can use its block when
4195 it is dynamic (useful for very large $message_headers, for example).
4197 6. Lookups now cache *every* query, not just the most recent. A new, separate
4198 store pool is used for this. It can be recovered when all lookup caches are
4199 flushed. Lookups now release memory at the end of their result strings.
4200 This has involved some general refactoring of the lookup sources.
4202 7. Some code has been added to the store_xxx() functions to reduce the amount
4203 of flapping under certain conditions.
4205 8. log_incoming_interface used to affect only the <= reception log lines. Now
4206 it causes the local interface and port to be added to several more SMTP log
4207 lines, for example "SMTP connection from", and rejection lines.
4209 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4211 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4213 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4214 could be overwritten at the end of the current message (or the start of a
4215 new message if it was set in a HELO ACL). The value is now preserved for
4216 the duration of the SMTP connection.
4218 12. If a transport had a headers_rewrite setting, and a matching header line
4219 contained an unqualified address, that address was qualified, even if it
4220 did not match any rewriting rules. The underlying bug was that the values
4221 of the flags that permit the existence of unqualified sender and recipient
4222 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4223 non-local messages, and by -bnq for local messages) were not being
4224 preserved with the message after it was received.
4226 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4227 "next input=" as part of the text comprising the host identity instead of
4228 the correct text. The code was using the same buffer for two different
4229 strings. However, depending on which order the printing function evaluated
4230 its arguments, the bug did not always show up. Under Linux, for example, my
4231 test suite worked just fine.
4233 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4234 doesn't work with some older versions of Perl. It has been changed to "my",
4235 which in any case is probably the better facility to use.
4237 15. A really picky compiler found some instances of statements for creating
4238 error messages that either had too many or two few arguments for the format
4241 16. The size of the buffer for calls to the DNS resolver has been increased
4242 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4243 for addresses that have a lot of PTR records. This alleviates a problem; it
4244 does not fully solve it.
4246 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4247 buffer now truncates the list and logs the incident, which is the same
4248 action as happens when Exim is looking up a host name and its aliases.
4249 Previously in this situation something unpredictable would happen;
4250 sometimes it was "internal error: store_reset failed".
4252 18. If a server dropped the connection unexpectedly when an Exim client was
4253 using GnuTLS and trying to read a response, the client delivery process
4254 crashed while trying to generate an error log message.
4256 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4257 string, for example:
4259 warn message = H1: something\nH2: something
4261 the text was added as a single header line from Exim's point of view
4262 though it ended up OK in the delivered message. However, searching for the
4263 second and subsequent header lines using $h_h2: did not work. This has been
4264 fixed. Similarly, if a system filter added multiple headers in this way,
4265 the routers could not see them.
4267 20. Expanded the error message when iplsearch is called with an invalid key to
4268 suggest using net-iplsearch in a host list.
4270 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4271 ACLs are no longer actually imposed (and a message to that effect is
4274 22. If a "gecos" field in a passwd entry contained escaped characters, in
4275 particular, if it contained a \" sequence, Exim got it wrong when building
4276 a From: or a Sender: header from that name. A second bug also caused
4277 incorrect handling when an unquoted " was present following a character
4278 that needed quoting.
4280 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4281 was not being matched caselessly.
4283 24. Arranged for all hyphens in the exim.8 source to be escaped with
4286 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4287 properly. Recipient callout cache records were still being keyed to include
4288 the sender, even when use_sender was set false. This led to far more
4289 callouts that were necessary. The sender is no longer included in the key
4290 when use_sender is false.
4292 26. Added "control = submission" modifier to ACLs.
4294 27. Added the ${base62d: operator to decode base 62 numbers.
4296 28. dnsdb lookups can now access SRV records.
4298 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4299 the configuration file.
4301 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4302 (-v) mode. This makes the output for a verbose queue run more intelligible.
4304 31. Added a use_postmaster feature to recipient callouts.
4306 32. Added the $body_zerocount variable, containing the number of binary zero
4307 bytes in the message body.
4309 33. The time of last modification of the "new" subdirectory is now used as the
4310 "mailbox time last read" when there is a quota error for a maildir
4313 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4315 35. Added +ignore_unknown as a special item in host lists.
4317 36. Code for decoding IPv6 addresses in host lists is now included, even if
4318 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4319 address was recognized as an IP address, but was then not correctly decoded
4320 into binary, causing unexpected and incorrect effects when compared with
4327 1. Very minor rewording of debugging text in manualroute to say "list of
4328 hosts" instead of "hostlist".
4330 2. If verify=header_syntax was set, and a header line with an unqualified
4331 address (no domain) and a large number of spaces between the end of the
4332 name and the colon was received, the reception process suffered a buffer
4333 overflow, and (when I tested it) crashed. This was caused by some obsolete
4334 code that should have been removed. The fix is to remove it!
4336 3. When running in the test harness, delay a bit after writing a bounce
4337 message to get a bit more predictability in the log output.
4339 4. Added a call to search_tidyup() just before forking a reception process. In
4340 theory, someone could use a lookup in the expansion of smtp_accept_max_
4341 per_host which, without the tidyup, could leave open a database connection.
4343 5. Added the variables $recipient_data and $sender_data which get set from a
4344 lookup success in an ACL "recipients" or "senders" condition, or a router
4345 "senders" option, similar to $domain_data and $local_part_data.
4347 6. Moved the writing of debug_print from before to after the "senders" test
4350 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4351 problems for message scanning, either using a data ACL, or using
4352 local_scan() because the Received: header was not generated till after they
4353 were called (in order to set the time as the time of reception completion).
4354 I have revised the way this works. The header is now generated after the
4355 body is received, but before the ACL or local_scan() are called. After they
4356 are run, the timestamp in the header is updated.
4362 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4363 before starting a queue runner without re-exec. This happened only when
4364 deliver_drop_privilege was set or when the Exim user was set to root. The
4365 effect of the bug was that timeouts during subsequent deliveries caused
4366 crashes instead of being properly handled. The handler is now left at its
4367 default (and expected) setting.
4369 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4370 message, again when deliver_drop_privilege is set or Exim is run as root.
4371 The bug described in (1) was not present in this case, but the tidying up
4372 of the other signals was missing. I have made the two cases consistent.
4374 3. The ignore_target_hosts setting on a manualroute router was being ignored
4375 for hosts that were looked up using the /MX notation.
4377 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4380 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4381 operated on the sender address. After changing the $sender_address to <>
4382 for the sender address verify, Exim was re-instated it as the original
4383 (before rewriting) address, but remembering that it had rewritten it, so it
4384 wasn't rewriting it again. This bug also had the effect of breaking the
4385 sender address verification caching when the sender address was rewritten.
4387 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4388 This has been changed so that if the ip literal address matches
4389 ignore_target_hosts, the router declines.
4391 7. Added expansion conditions match_domain, match_address, and match_local_
4392 part (NOT match_host).
4394 8. The placeholder for the Received: header didn't have a length field set.
4396 9. Added code to Exim itself and to exim_lock to test for a specific race
4397 condition that could lead to file corruption when using MBX delivery. The
4398 issue is with the lockfile that is created in /tmp. If this file is removed
4399 after a process has opened it but before that process has acquired a lock,
4400 there is the potential for a second process to recreate the file and also
4401 acquire a lock. This could lead to two Exim processes writing to the file
4402 at the same time. The added code performs the same test as UW imapd; it
4403 checks after acquiring the lock that its file descriptor still refers to
4404 the same named file.
4406 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4407 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4410 11. Added the smtp_active_hostname option. If used, this will typically be made
4411 to depend on the incoming interface address. Because $interface_address is
4412 not set up until the daemon has forked a reception process, error responses
4413 that can happen earlier (such as "too many connections") no longer contain
4416 12. If an expansion in a condition on a "warn" statement fails because a lookup
4417 defers, the "warn" statement is abandoned, and the next ACL statement is
4418 processed. Previously this caused the whole ACL to be aborted.
4420 13. Added the iplsearch lookup type.
4422 14. Added ident_timeout as a log selector.
4424 15. Added tls_certificate_verified as a log selector.
4426 16. Added a global option tls_require_ciphers (compare the smtp transport
4427 option of the same name). This controls incoming TLS connections.
4429 17. I finally figured out how to make tls_require_ciphers do a similar thing
4430 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
4431 before starting the TLS session.
4433 18. Tabs are now shown as \t in -bP output.
4435 19. If the log selector return_path_on_delivery was set, Exim crashed when
4436 bouncing a message because it had too many Received: header lines.
4438 20. If two routers both had headers_remove settings, and the first one included
4439 a superfluous trailing colon, the final name in the first list and the
4440 first name in the second list were incorrectly joined into one item (with a
4441 colon in the middle).
4447 1. Added -C and -D options to the exinext utility, mainly to make it easier
4448 to include in the automated testing, but these could be helpful when
4449 multiple configurations are in use.
4451 2. The exinext utility was not formatting the output nicely when there was
4452 an alternate port involved in the retry record key, nor when there was a
4453 message id as well (for retries that were specific to a specific message
4454 and a specific host). It was also confused by IPv6 addresses, because of
4455 the additional colons they contain. I have fixed the IPv4 problem, and
4456 patched it up to do a reasonable job for IPv6.
4458 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
4459 delivery, the log line now contains "pipelined" if PIPELINING was used.
4461 4. An SMTP transport process used to panic and die if the bind() call to set
4462 an explicit outgoing interface failed. This has been changed; it is now
4463 treated in the same way as a connect() failure.
4465 5. A reference to $sender_host_name in the part of a conditional expansion
4466 that was being skipped was still causing a DNS lookup. This no longer
4469 6. The def: expansion condition was not recognizing references to header lines
4470 that used bh_ and bheader_.
4472 7. Added the _cache feature to named lists.
4474 8. The code for checking quota_filecount in the appendfile transport was
4475 allowing one more file than it should have been.
4477 9. For compatibility with Sendmail, the command line option
4485 and sets the incoming protocol and host name (for trusted callers). The
4486 host name and its colon can be omitted when only the protocol is to be set.
4487 Note the Exim already has two private options, -pd and -ps, that refer to
4488 embedded Perl. It is therefore impossible to set a protocol value of "d" or
4489 "s", but I don't think that's a major issue.
4491 10. A number of refactoring changes to the code, none of which should affect
4494 (a) The number of logging options was getting close to filling up the
4495 32-bit word that was used as a bit map. I have split them into two classes:
4496 those that are passed in the argument to log_write(), and those that are
4497 only ever tested independently outside of that function. These are now in
4498 separate 32-bit words, so there is plenty of room for expansion again.
4499 There is no change in the user interface or the logging behaviour.
4501 (b) When building, for example, log lines, the code previously used a
4502 macro that called string_cat() twice, in order to add two strings. This is
4503 not really sufficiently general. Furthermore, there was one instance where
4504 it was actually wrong because one of the argument was used twice, and in
4505 one call a function was used. (As it happened, calling the function twice
4506 did not affect the overall behaviour.) The macro has been replaced by a
4507 function that can join an arbitrary number of extra strings onto a growing
4510 (c) The code for expansion conditions now uses a table and a binary chop
4511 instead of a serial search (which was left over from when there were very
4512 few conditions). Also, it now recognizes conditions like "pam" even when
4513 the relevant support is not compiled in: a suitably worded error message is
4514 given if an attempt is made to use such a condition.
4516 11. Added ${time_interval:xxxxx}.
4518 12. A bug was causing one of the ddress fields not to be passed back correctly
4519 from remote delivery subprocesses. The field in question was not being
4520 subsequently used, so this caused to problems in practice.
4522 13. Added new log selectors queue_time and deliver_time.
4524 14. Might have fixed a bug in maildirsizefile handling that threw up
4525 "unexpected character" debug warnings, and recalculated the data
4526 unnecessarily. In any case, I expanded the warning message to give more
4529 15. Added the message "Restricted characters in address" to the statements in
4530 the default ACL that block characters like @ and % in local parts.
4532 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
4533 Three changes have been made:
4535 (a) There was a serious bug; a negative response to MAIL caused the whole
4536 recipient domain to be cached as invalid, thereby blocking all messages
4537 to all local parts at the same domain, from all senders. This bug has
4538 been fixed. The domain is no longer cached after a negative response to
4539 MAIL if the sender used is not empty.
4541 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
4544 (c) A new callout option, "use_sender" has been added for people who want
4545 the modified behaviour.
4551 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
4554 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
4555 indeed breaks things for older releases.
4557 3. Added additional logging to the case where there is a problem reading data
4558 from a filter that is running in a subprocess using a pipe, in order to
4559 try to track down a specific problem.
4561 4. Testing facility fudge: when running in the test harness and attempting
4562 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
4563 getting "No route to host". Convert this to a timeout.
4565 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
4568 6. Some OS don't have socklen_t but use size_t instead. This affects the
4569 fifth argument of getsockopt() amongst other things. This is now
4570 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
4571 can be set for individual OS. I have set it for SunOS5, OSF1, and
4572 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
4573 some earlier ones do not.
4575 7. Change 4.30/15 was not doing the test caselessly.
4577 8. The standard form for an IPv6 address literal was being rejected by address
4578 parsing in, for example, MAIL and RCPT commands. An example of this kind of
4579 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4580 this, as well as the form without the "IPv6" on the front (but only when
4581 address literals are enabled, of course).
4583 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
4585 10. Exim crashed if a message with an empty sender address specified by -f
4586 encountered a router with an errors_to setting. This could be provoked only
4587 by a command such as
4591 where an empty string was supplied; "<>" did not hit this bug.
4593 11. Installed PCRE release 4.5.
4595 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
4596 remained set. It is now erased.
4598 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
4599 times from message ids (which are base 36 rather than the normal 62).
4601 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
4602 were being counted as actual protocol errors, and logged if the log
4603 selector +smtp_protocol_error was set. One cannot be perfect in this test,
4604 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
4605 and DATA following a set of rejected RCPTs do not count as protocol errors.
4606 In other words, Exim assumes they were pipelined, though this may not
4607 actually be the case. Of course, in all cases the client gets an
4608 appropriate error code.
4610 15. If a lookup fails in an ACL condition, a message about the failure may
4611 be available; it is used if testing the ACL cannot continue, because most
4612 such messages specify what the cause of the deferral is. However, some
4613 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
4614 that caused an old message to be retained and used if a later statement
4615 caused a defer, replacing the real cause of the deferral.
4617 16. If an IP address had so many PTR records that the DNS lookup buffer
4618 was not large enough to hold them, Exim could crash while trying to process
4619 the truncated data. It now detects and logs this case.
4621 17. Further to 4.21/58, another change has been made: if (and only if) the
4622 first line of a message (the first header line) ends with CRLF, a bare LF
4623 in a subsequent header line has a space inserted after it, so as not to
4624 terminate the header.
4626 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
4627 unnecessarily, used atoi() instead of strtol(), and didn't check the
4628 termination when getting file sizes from file names by regex.
4630 19. Completely re-implemented the support for maildirsize files, in the light
4631 of a number of problems with the previous contributed implementation
4632 (4.30/29). In particular:
4634 . If the quota is zero, the maildirsize file is maintained, but no quota is
4637 . If the maildir directory does not exist, it is created before any attempt
4638 to write a maildirsize file.
4640 . The quota value in the file is just a cache; if the quota is changed in
4641 the transport, the new value overrides.
4643 . A regular expression is available for excluding directories from the
4646 20. The autoreply transport checks the characters in options that define the
4647 message's headers; it allows continued headers, but it was checking with
4648 isspace() after an embedded newline instead of explicitly looking for a
4651 21. If all the "regular" hosts to which an address was routed had passed their
4652 expiry times, and had not reached their retry times, the address was
4653 bounced, even if fallback hosts were defined. Now Exim should go on to try
4656 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
4657 equivalent code in the SMTP transport. Some hosts send humungous responses
4658 to HELO/EHLO, more than 1024 it seems.
4660 23. Refactoring: code in filter.c used (void *) for "any old type" but this
4661 gives compiler warnings in some environments. I've now done it "properly",
4664 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
4665 (because of problems with the built-in one) was declared to return uschar *
4666 instead of char *, causing compiler failure.
4668 25. Fixed a file descriptor leak when processing alias/forward files.
4670 26. Fixed a minor format string issue in dbfn.c.
4672 27. Typo in exim.c: ("dmbnz" for "dbmnz").
4674 28. If a filter file refered to $h_xxx or $message_headers, and the headers
4675 contained RFC 2047 "words", Exim's memory could, under certain conditions,
4678 29. When a sender address is verified, it is cached, to save repeating the test
4679 when there is more than one recipient in a message. However, when the
4680 verification involves a callout, it is possible for different callout
4681 options to be set for different recipients. It is too complicated to keep
4682 track of this in the cache, so now Exim always runs a verification when a
4683 callout is required, relying on the callout cache for the optimization.
4684 The overhead is duplication of the address routing, but this should not be
4687 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
4688 to be verified with callout=postmaster, and the main callout worked but the
4689 postmaster check failed, the verification correctly failed. However, if a
4690 subsequent RCPT command asked for sender verification *without* the
4691 postmaster check, incorrect caching caused this verification also to fail,
4694 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
4695 it was not caching the DNS options (qualify_single, search_parents) that
4696 were used when the lookup failed. A subsequent lookup with different
4697 options therefore always gave the same answer, though there were cases
4698 where it should not have. (Example: a "domains = !$mx_any" option on a
4699 dnslookup router: the "domains" option is always processed without any
4700 widening, but the router might have qualify_single set.) Now Exim uses the
4701 cached value only when the same options are set.
4703 32. Added John Jetmore's "exipick" utility to the distribution.
4705 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
4706 than a timeout (e.g. a certificate is required, and is not provided), an
4707 Exim server now closes the connection immediately. Previously it waited for
4708 the client to close - but if the client is SSL, it seems that they each
4709 wait for each other, leading to a delay before one of them times out.
4711 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
4712 maintained 0.8.x compatibility because I don't think many are using it, and
4713 it is clearly obsolete.
4715 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
4718 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
4719 client certificate was expired. A simple patch fixes this, though I don't
4720 understand the full logic of why the verify callback is called multiple
4723 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
4724 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
4725 which causes problems with some clients (such as the Certicom SSL Plus
4726 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
4727 disables the coutermeasure allowing Eudora to connect."
4729 38. Exim was not checking that a write() to a log file succeeded. This could
4730 lead to Bad Things if a log got too big, in particular if it hit a file
4731 size limit. Exim now panics and dies if it cannot write to a log file, just
4732 as it does if it cannot open a log file.
4734 39. Modified OS/Makefile-Linux so that it now contains
4736 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
4738 The two -D definitions ensure that Exim is compiled with large file
4739 support, which makes it possible to handle log files that are bigger than
4742 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
4743 instance) a domain was checked against a named list that involved a lookup,
4744 causing $domain_data to be set, then another domain was checked against the
4745 same list, then the first domain was re-checked, the value of $domain_data
4746 after the final check could be wrong. In particular, if the second check
4747 failed, it could be set empty. This bug probably also applied to
4750 41. The strip_trailing_dot option was not being applied to the address given
4751 with the -f command-line option.
4753 42. The code for reading a message's header from the spool was incrementing
4754 $received_count, but never initializing it. This meant that the value was
4755 incorrect (doubled) while delivering a message in the same process in which
4756 it was received. In the most common configuration of Exim, this never
4757 happens - a fresh exec is done - but it can happen when
4758 deliver_drop_privilege is set.
4760 43. When Exim logs an SMTP synchronization error - client data sent too soon -
4761 it now includes up to 150 characters of the unexpected data in the log
4764 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
4765 and building data strings. The size of both of these buffers was 10 000
4766 bytes - far larger than anybody would *ever* want, thought I. Needless to
4767 say, somebody hit the limit. I have increased the maximum line length to
4768 20 000 and the maximum data length of concatenated lines to 100 000. I have
4769 also fixed two bugs, because there was no checking on these buffers. Tsk,
4770 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
4771 buffer is too small.
4773 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
4774 lsearch lookups. Now it does.
4776 46. When parsing a route_list item in a manualroute router, a fixed-length
4777 buffer was used for the list of hosts. I made this 1024 bytes long,
4778 thinking that nobody would ever have a list of hosts that long. Wrong.
4779 Somebody had a whole pile of complicated expansion conditions, and the
4780 string was silently truncated, leading to an expansion error. It turns out
4781 that it is easier to change to an unlimited length (owing to other changes
4782 that have happened since this code was originally written) than to build
4783 structure for giving a limitation error. The length of the item that
4784 expands into the list of hosts is now unlimited.
4786 47. The lsearch lookup could not handle data where the length of text line was
4787 more than 4095 characters. Such lines were truncated, leading to shortened
4788 data being returned. It should now handle lines of any length.
4790 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
4791 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
4794 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
4795 replacing strings such as BIN_DIRECTORY in a source file: the replacement
4796 no longer happens in comment lines. A list of replacements is now placed
4797 at the head of all of the source files, except those whose only change is
4798 to replace PERL_COMMAND in the very first #! line.
4800 50. Replaced the slow insertion sort in queue.c, for sorting the list of
4801 messages on the queue, with a bottom-up merge sort, using code contributed
4802 by Michael Haardt. This should make operations like -bp somewhat faster on
4803 large queues. It won't affect queue runners, except when queue_run_in_order
4806 51. Installed eximstats 1.31 in the distribution.
4808 52. Added support for SRV lookups to the dnslookup router.
4810 53. If an ACL referred to $message_body or $message_body_end, the value was not
4811 reset for any messages that followed in the same SMTP session.
4813 54. The store-handling optimization for building very long strings was not
4814 differentiating between the different store pools. I don't think this
4815 actually made any difference in practice, but I've tidied it.
4817 55. While running the routers to verify a sender address, $sender_address
4818 was still set to the sender address. This is wrong, because when routing to
4819 send a bounce to the sender, it would be empty. Therefore, I have changed
4820 it so that, while verifying a sender address, $sender_address is set to <>.
4821 (There is no change to what happens when verifying a recipient address.)
4823 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
4824 target A or AAAA records (if not already returned) without resetting the
4825 qualify_single or search_parents options of the DNS resolver. These are
4826 inappropriate in this case because the targets of MX and SRV records must
4827 be FQDNs. A broken DNS record could cause trouble if it happened to have a
4828 target that, when qualified, matched something in the local domain. These
4829 two options are now turned off when doing these lookups.
4831 57. It seems that at least some releases of Reiserfs (which does not have the
4832 concept of a fixed number of inodes) returns zero and not -1 for the
4833 number of available inodes. This interacted badly with check_spool_inodes,
4834 which assumed that -1 was the "no such thing" setting. What I have done is
4835 to check that the total number of inodes is greater than zero before doing
4836 the test of how many are available.
4838 58. When a "warn" ACL statement has a log_message modifier, the message is
4839 remembered, and not repeated. This is to avoid a lot of repetition when a
4840 message has many recipients that cause the same warning to be written.
4841 Howewer, Exim was preserving the list of already written lines for an
4842 entire SMTP session, which doesn't seem right. The memory is now reset if a
4843 new message is started.
4845 59. The "rewrite" debugging flag was not showing the result of rewriting in the
4846 debugging output unless log_rewrite was also set.
4848 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
4849 of (int)(handle) when we know that handle contains (void *)(-1).
4851 61. The Exim daemon panic-logs an error return when it closes the incoming
4852 connection. However "connection reset by peer" seems to be common, and
4853 isn't really an error worthy of noting specially, so that particular error
4856 62. When Exim is trying to find all the local interfaces, it used to panic and
4857 die if the ioctl to get the interface flags failed. However, it seems that
4858 on at least one OS (Solaris 9) it is possible to have an interface that is
4859 included in the list of interfaces, but for which you get a failure error
4860 for this call. This happens when the interface is not "plumbed" into a
4861 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
4862 failure of the "get flags" call assumes that the interface is down.
4864 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
4865 makes life easier for people who are doing arithmetic on fields extracted
4866 from dates, where you often get leading zeros that should not be
4867 interpreted as octal.
4869 64. Added qualify_domain to the redirect router, to override the global
4872 65. If a pathologically long header line contained very many addresses (the
4873 report of this problem mentioned 10 000) and each of them was rewritten,
4874 Exim could use up a very large amount of memory. (It kept on making new
4875 copies of the header line as it rewrote, and never released the old ones.)
4876 At the expense of a bit more processing, the header rewriting function has
4877 been changed so that it no longer eats memory in this way.
4879 66. The generation of the Received: header has been moved from the time that a
4880 message starts to be received, to the time that it finishes. The timestamp
4881 in the Received: header should now be very close to that of the <= log
4882 line. There are two side-effects of this change:
4884 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
4885 logged header lines no longer include the local Received: line, because
4886 it has not yet been created. The same applies to a copy of the message
4887 that is returned to a non-SMTP sender when a message is rejected.
4889 (b) When a filter file is tested using -bf, no additional Received: header
4890 is added to the test message. After some thought, I decided that this
4893 This change does not affect the value of $received_for. It is still set
4894 after address rewriting, but before local_scan() is called.
4896 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
4898 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
4899 gave an unhelpful panic error message, and a defer error. I have managed to
4900 change this behaviour so that it now rejects any supplied certificate,
4901 which seems right, as the list of acceptable certificates is empty.
4903 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
4904 gave an unhelpful defer error. I have not managed to make this reject any
4905 supplied certificates, but the error message it gives is "no certificate
4906 supplied", which is not helpful.
4908 70. exigrep's output now also includes lines that are not associated with any
4909 message, but which match the given pattern. Implemented by a patch from
4910 Martin Sluka, which also tidied up the Perl a bit.
4912 71. Recipient callout verification, like sender verification, was using <> in
4913 the MAIL FROM command. This isn't really the right thing, since the actual
4914 sender may affect whether the remote host accepts the recipient or not. I
4915 have changed it to use the actual sender in the callout; this means that
4916 the cache record is now keyed on a recipient/sender pair, not just the
4917 recipient address. There doesn't seem to be a real danger of callout loops,
4918 since a callout by the remote host to check the sender would use <>.
4919 [SEE ABOVE: changed after hitting problems.]
4921 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
4922 temporary errors. However, in the case of such a code being given after
4923 the end of a data transmission (i.e. after ".") Exim was failing to write
4924 a retry record for the message. (Yes, there was some broken host that was
4925 actually sending 8xx at this point.)
4927 73. An unknown lookup type in a host list could cause Exim to panic-die when
4928 the list was checked. (An example that provoked this was putting <; in the
4929 middle of a list instead of at the start.) If this happened during a DATA
4930 ACL check, a -D file could be left lying around. This kind of configuration
4931 error no longer causes Exim to die; instead it causes a defer errror. The
4932 incident is still logged to the main and panic logs.
4934 74. Buglet left over from Exim 3 conversion. The message "too many messages
4935 in one connection" was written to the rejectlog but not the mainlog, except
4936 when address rewriting (yes!) was being logged.
4938 75. Added write_rejectlog option.
4940 76. When a system filter was run not as root (that is, when system_filter_user
4941 was set), the values of the $n variables were not being returned to the
4942 main process; thus, they were not subsequently available in the $sn
4945 77. Added +return_path_on_delivery log selector.
4947 78. A connection timeout was being treated differently from recipients deferred
4948 when testing hosts_max_try with a message that was older than the host's
4949 retry timeout. (The host should not be counted, thus allowing all hosts to
4950 be tried at least once before bouncing.) This may have been the cause of an
4951 occasionally reported bug whereby a message would remain on the queue
4952 longer than the retry timeout, but would be bounced if a delivery was
4953 forced. I say "may" because I never totally pinned down the problem;
4954 setting up timeout/retry tests is difficult. See also the next item.
4956 79. The ultimate address timeout was not being applied to errors that involved
4957 a combination of host plus message (for example, a timeout on a MAIL
4958 command). When an address resolved to a number of possible hosts, and they
4959 were not all tried for each delivery (e.g. because of hosts_max_try), a
4960 message could remain on the queue longer than the retry timeout.
4962 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4965 81. Fixed an obscure SMTP outgoing bug which required at least the following
4966 conditions: (a) there was another message waiting for the same server;
4967 (b) the server returned 5xx to all RCPT commands in the first message so
4968 that the message was not completed; (c) the server dropped the connection
4969 or gave a negative response to the RSET that Exim sends to abort the
4970 transaction. The observed case was a dropped connection after DATA that had
4971 been sent in pipelining mode. That is, the server had advertised PIPELINING
4972 but was not implementing it correctly. The effect of the bug was incorrect
4973 behaviour, such as trying another host, and this could lead to a crash.
4979 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4980 and daemon.c were passed as pointers to ints; they should have been
4981 pointers to socklen_t variables (which are typically unsigned ints).
4983 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4986 3. Fixed a really odd bug that affected only the testing scheme; patching a
4987 certain fixed string in the binary changed the value of another string that
4988 happened to be identical to the end of the original first string.
4990 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4991 name", it returns that address as the IP address. On some operating
4992 systems (e.g. Solaris), it also passes back the IP address string as the
4993 "host name". However, on others (e.g. Linux), it passes back an empty
4994 string. Exim wasn't checking for this, and was changing the host name to an
4995 empty string, assuming it had been canonicized.
4997 5. Although rare, it is permitted to have more than one PTR record for a given
4998 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4999 all the names associated with an address, because they do in Solaris.
5000 However, it seems that they do not in Linux for data that comes from the
5001 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5002 I found this out when I moved to a new Linux workstation and tried to run
5003 the Exim test suite.
5005 To get round this problem I have changed the code so that it now does its
5006 own call to the DNS to look up PTR records when searching for a host name.
5007 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5008 addresses that are only in /etc/hosts are still found.
5010 This behaviour is, however, controlled by an option called host_lookup_
5011 order, which defaults to "bydns:byaddr". If people want to use the other
5012 order, or indeed, just use one or the other means of lookup, they can
5013 specify it in this variable.
5015 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5016 some operating systems, this comes back from gethostbyaddr() as an empty
5017 string, and this is what Exim used to test for. However, it seems that in
5018 other systems, "." is yielded. Exim now tests for this case too.
5020 7. The values of check_spool_space and check_log_space are now held internally
5021 as a number of kilobytes instead of an absolute number of bytes. If a
5022 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5023 kilobyte. This means that much larger values can be stored.
5025 8. Exim monitor: an attempt to get the action menu when not actually pointing
5026 at a message produces an empty menu entitled "No message selected". This
5027 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5028 no entries in it ("Shell widget menu has zero width and/or height"). So I
5029 have added a single, blank menu entry in this case.
5031 9. Added ${quote_local_part.
5033 10. MIME decoding is now applied to the contents of Subject: header lines when
5036 11. Now that a reference to $sender_host_address automatically causes a reverse
5037 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5038 host lookup before query-style lookups in lists that might use this
5039 variable. This has therefore been abolished, and the "net-" prefix is no
5040 longer necessary for query-style lookups.
5042 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5043 have been a typo for LFLAGS, so it has been changed.
5045 13. The install script calls Exim with "-C /dev/null" in order to find the
5046 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5047 to be output. Howeve, since Exim outputs its version number before the
5048 error, it didn't break the script. It just looked ugly. I fixed this by
5049 always allowing "-C /dev/null" if the caller is root.
5051 14. Ignore overlarge ACL variable number when reading spool file - insurance
5052 against a later release with more variables having written the file.
5054 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5055 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5056 this, as well as the form without the "IPv6" on the front.
5058 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5059 OS/Makefile-Darwin file.
5061 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5062 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5064 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5065 variable. However, some OS use ints for inodes. I've added cast to long int
5066 to get rid of the compiler warning.
5068 19. I had forgotten to lock out "/../" in configuration file names when
5069 ALT_CONFIG_PREFIX was set.
5071 20. Routers used for verification do not need to specify transports. However,
5072 if such a router generated a host list, and callout was configured, Exim
5073 crashed, because it could not find a port number from the (non-existent)
5074 transport. It now assumes port 25 in this circumstance.
5076 21. Added the -t option to exigrep.
5078 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5079 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5080 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5081 them is set, LOOKUP_LSEARCH is forced.
5083 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5085 24. Added sender and host information to the "rejected by local_scan()" log
5086 line; previously there was no indication of these.
5088 25. Added .include_if_exists.
5090 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5091 Linux. It turns out that not all file systems support this. Apparently some
5092 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5093 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5094 not supported on the file descriptor, is now ignored when Exim is trying to
5095 sync a directory. This applies only to Linux.
5097 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5099 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5100 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5103 29. Added support for maildirsize files from supplied patch (modified a bit).
5105 30. The use of :fail: followed by an empty string could lead Exim to respond to
5106 sender verification failures with (e.g.):
5108 550 Verification failed for <xxx>
5109 550 Sender verify failed
5111 where the first response line was missing the '-' that indicates it is not
5112 the final line of the response.
5114 31. The loop for finding the name of the user that called Exim had a hardwired
5115 limit of 10; it now uses the value of finduser_retries, which is used for
5116 all other user lookups.
5118 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5121 33. Exim was neglecting to zero errno before one call of strtol() when
5122 expanding a string and expecting an integer value. On some systems this
5123 resulted in spurious "integer overflow" errors. Also, it was casting the
5124 result into an int without checking.
5126 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5127 did not work. The code looks as if it has *never* worked, though it appears
5128 to have been documented since at least releast 1.62. I have made it work.
5130 35. The "timeout_DNS" error in retry rules, also documented since at least
5131 1.62, also never worked. As it isn't clear exactly what this means, and
5132 clearly it isn't a major issue, I have abolished the feature by treating it
5133 as "timeout", and writing a warning to the main and panic logs.
5135 36. The display of retry rules for -brt wasn't always showing the error code
5138 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5139 timeout_connect_A, timeout_connect_MX.
5141 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5142 to the empty sender.
5144 39. The daemon was not analysing the content of -oX till after it had closed
5145 stderr and disconnected from the controlling terminal. This meant that any
5146 syntax errors were only noted on the panic log, and the return code from
5147 the command was 0. By re-arranging the code a little, I've made the
5148 decoding happen first, so such errors now appear on stderr, and the return
5149 code is 1. However, the actual setting up of the sockets still happens in
5150 the disconnected process, so errors there are still only recorded on the
5153 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5154 connections (as happens on some IP stacks) was logged at start up time as
5155 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5156 it from "IPv6 and IPv4", which means that two separate sockets are being
5159 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5160 says whether AF_INET or AF_INET6 was passed as an argument.
5162 42. Exiwhat output was messed up when time zones were included in log
5165 43. Exiwhat now gives more information about the daemon's listening ports,
5166 and whether -tls-on-connect was used.
5168 44. The "port" option of the smtp transport is now expanded.
5170 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5171 silently ignored. Now an error message is written to the main and panic
5174 46. There's a new ACL modifier called "logwrite" which writes to a log file
5175 as soon as it is encountered.
5177 47. Added $local_user_uid and $local_user_gid at routing time.
5179 48. Exim crashed when trying to verify a sender address that was being
5182 49. Exim was recognizing only a space character after ".include". It now also
5183 recognizes a tab character.
5185 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5186 extracting the relevant information from the specification. The man page no
5187 longer contains scrambled data for the -d option, and I've added a section
5188 at the front about calling Exim under different names.
5190 51. Added "extra_headers" argument to the "mail" command in filter files.
5192 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5195 53. Installed eximstats 1.29.
5197 54. Added transport_filter_timeout as a generic transport option.
5199 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5200 Cc: header lines. This was required by RFC 822, but it not required by RFC
5203 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5204 incoming messages that did not have them. Now it does so only if the
5205 message originates locally, that is, if there is no associated remote host
5206 address. When Resent- header lines are present, this applies to the Resent-
5207 lines rather than the non-Resent- lines.
5209 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5210 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5212 58. Messages for configuration errors now include the name of the main
5213 configuration file - useful now that there may be more than one file in a
5214 list (.included file names were always shown).
5216 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5217 for those rare installations that do not start the daemon as root or run it
5218 setuid root. I've cut out the call to initgroups() if the daemon is not
5221 60. The Exim user and group can now be bound into the binary as text strings
5222 that are looked up at the start of Exim's processing.
5224 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5226 62. Added $mailstore_basename variable.
5228 63. Installed patch to sieve.c from Michael Haardt.
5230 64. When Exim failed to open the panic log after failing to open the main log,
5231 the original message it was trying to log was written to stderr and debug
5232 output, but if they were not available (the usual case in production), it
5233 was lost. Now it is written to syslog before the two lines that record the
5234 failures to open the logs.
5236 65. Users' Exim filters run in subprocesses under the user's uid. It is
5237 possible for a "deliver" command or an alias in a "personal" command to
5238 provoke an address rewrite. If logging of address rewriting is configured,
5239 this fails because the process is not running as root or exim. There may be
5240 a better way of dealing with this, but for the moment (because 4.30 needs
5241 to be released), I have disabled address rewrite logging when running a
5242 filter in a non-root, non-exim process.
5248 1. The buildconfig auxiliary program wasn't quoting the value set for
5249 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5250 not defined. This bug was masked in 4.22 by the effect that was fixed in
5253 2. Some messages that were rejected after a message id was allocated were
5254 shown as "incomplete" by exigrep. It no longer does this for messages that
5255 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5257 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5258 have allow_domain_literals set, the ID did not get logged in the <= line.
5259 Domain literals are now always recognized in Message-ID: header lines.
5261 4. The first argument for a ${extract expansion item is the key name or field
5262 number. Leading and trailing spaces in this item were not being ignored,
5263 causing some misleading effects.
5265 5. When deliver_drop_privilege was set, single queue runner processes started
5266 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5267 same command in the process it spins off) were not dropping privilege.
5269 6. When the daemon running as "exim" started a queue runner, it always
5270 re-executed Exim in the spun-off process. This is a waste of effort when
5271 deliver_drop_privilege is set. The new process now just calls the
5272 queue-runner function directly.
5278 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5281 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5282 ignored. Though the use of -oP was forcing the writing of a pid file, it
5283 was always written to the default place.
5285 3. If the message "no IP address found for host xxxx" is generated during
5286 incoming verification, it is now followed by identification of the incoming
5287 connection (so you can more easily find what provoked it).
5289 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5291 5. Added some features to "harden" Exim a bit more against certain attacks:
5293 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5294 be put in Local/Makefile. This is like the never_users runtime option,
5295 but it cannot be overridden. The default setting is "root".
5297 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5298 prefix string with which any file named in a -C command line option
5301 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5302 is retained for -C and -D only if the caller of Exim is root. Without
5303 it, the exim user may also use -C and -D and retain privilege.
5305 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5306 command line option is disabled.
5308 6. Macro names set by the -D option must start with an upper case letter, just
5309 like macro names defined in the configuration file.
5311 7. Added "dereference=" facility to LDAP.
5313 8. Two instances of the typo "uknown" in the source files are fixed.
5315 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5316 the Configure-Makefile script screwed up while processing it.
5318 10. Incorporated PCRE 4.4.
5320 11. The SMTP synchronization check was not operating right at the start of an
5321 SMTP session. For example, it could not catch a HELO sent before the client
5322 waited for the greeting. There is now a check for outstanding input at the
5323 point when the greeting is written. Because of the duplex, asynchronous
5324 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5325 way, but not yet received, when the check is performed.
5327 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5328 on TCP/IP sockets, because this apparently causes some broken clients to
5331 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5332 unchanged) from the Cygwin maintainer.
5334 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5335 was supported instead of testing for mbx. Effectively a typo.
5337 15. The spa authenticator server code was not checking that the input it
5338 received was valid base64.
5340 16. The debug output line for the "set" modifier in ACLs was not showing the
5341 name of the variable that was being set.
5343 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5345 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5346 lookup on its own. Something else was needed to trigger the lookup. For
5347 example, a match in host_lookup or the need for a host name in a host list.
5348 Now, if $sender_host_name is referenced and the host name has not yet been
5349 looked up, a lookup is performed. If the lookup fails, the variable remains
5350 empty, and $host_lookup_failed is set to "1".
5352 19. Added "eqi" as a case-independent comparison operator.
5354 20. The saslauthd authentication condition could segfault if neither service
5355 nor realm was specified.
5357 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5358 error message that was logged was misleading, and incoming SMTP
5359 connections were dropped. The message is now more accurate, and temporary
5360 errors are given to SMTP connections.
5362 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5363 (except RSET) with a 421 error, until QUIT is received. However, it was
5364 failing to send a response to QUIT.
5366 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5367 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5368 ACL is now run after the helo_try_verify_hosts code.
5370 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5371 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5372 case-independent, so other case variants are also recognized). Apparently
5373 some systems use these upper case variants.
5375 25. If more than two messages were waiting for the same host, and a transport
5376 filter was specified for the transport, Exim sent two messages over the
5377 same TCP/IP connection, and then failed with "socket operation on non-
5378 socket" when it tried to send the third.
5380 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5382 27. The extern definition of crypt16() in expand.c was not being excluded when
5383 the OS had its own crypt16() function.
5385 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5386 as a preferred synonym for return_size_limit, both as an option and as an
5389 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5391 30. Changed the default configuration ACL to relax the local part checking rule
5392 for addresses that are not in any local domains. For these addresses,
5393 slashes and pipe symbols are allowed within local parts, but the sequence
5394 /../ is explicitly forbidden.
5396 31. SPA server authentication was not clearing the challenge buffer before
5399 32. log_message in a "warn" ACL statement was writing to the reject log as
5400 well as to the main log, which contradicts the documentation and doesn't
5401 seem right (because no rejection is happening). So I have stopped it.
5403 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5404 However, I am unable to do any testing of this.
5406 34. Fixed an infelicity in the appendfile transport. When checking directories
5407 for a mailbox, to see if any needed to be created, it was accidentally
5408 using path names with one or more superfluous leading slashes; tracing
5409 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5411 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5412 contents are added to the log line that is written for every discarded
5413 recipient. (Previously a log_message setting was ignored.)
5415 36. The ${quote: operator now quotes the string if it is empty.
5417 37. The install script runs exim in order to find its version number. If for
5418 some reason other than non-existence or emptiness, which it checks, it
5419 could not run './exim', it was installing it with an empty version number,
5420 i.e. as "exim-". This error state is now caught, and the installation is
5423 38. An argument was missing from the function that creates an error message
5424 when Exim fails to connect to the socket for saslauthd authentication.
5425 This could cause Exim to crash, or give a corrupted message.
5427 39. Added isip, isip4, and isip6 to ${if conditions.
5429 40. The ACL variables $acl_xx are now saved with the message, and can be
5430 accessed later in routers, transports, and filters.
5432 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
5433 strings in the file are not string-expanded.
5435 42. If a MAIL command specified a SIZE value that was too large to fit into an
5436 int variable, the check against message_size_limit failed. Such values are
5437 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
5438 day this will have to be increased, but I don't think I want to be around
5439 when emails are that large.
5446 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
5447 iconv() is not standard in FreeBSD.
5449 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
5450 IPv6 enabled. The observed symptom was a segmentation fault on return from
5451 the function os_common_find_running_interfaces() in src/os.c.
5453 3. In the check_special_case() function in daemon.c I had used "errno" as an
5454 argument name, which causes warnings on some systems. This was basically a
5455 typo, since it was named "eno" in the comments!
5457 4. The code that waits for the clock to tick (at a resolution of some fraction
5458 of a second) so as to ensure message-id uniqueness was always waiting for
5459 at least one whole tick, when it could have waited for less. [This is
5460 almost certainly not relevant at current processor speeds, where it is
5461 unlikely to ever wait at all. But we try to future-proof.]
5463 5. The function that sleeps for a time interval that includes fractions of a
5464 second contained a race. It did not block SIGALRM between setting the
5465 timer, and suspending (a couple of lines later). If the interval was short
5466 and the sigsuspend() was delayed until after it had expired, the suspension
5467 never ended. On busy systems this could lead to processes getting stuck for
5470 6. Some uncommon configurations may cause a lookup to happen in a queue runner
5471 process, before it forks any delivery processes. The open lookup caching
5472 mechanism meant that the open file or database connection was passed into
5473 the delivery process. The problem was that delivery processes always tidy
5474 up cached lookup data. This could cause a problem for the next delivery
5475 process started by the queue runner, because the external queue runner
5476 process does not know about the closure. So the next delivery process
5477 still has data in the lookup cache. In the case of a file lookup, there was
5478 no problem because closing a file descriptor in a subprocess doesn't affect
5479 the parent. However, if the lookup was caching a connection to a database,
5480 the connection was closed, and the second delivery process was likely to
5481 see errors such as "PGSQL: query failed: server closed the connection
5482 unexpectedly". The problem has been fixed by closing all cached lookups
5483 in a queue runner before running a delivery process.
5485 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
5486 seem to have the "const" qualifier which it has on other OS. I've
5489 8. Change 4.21/2 was too strict. It is only if there are two authenticators
5490 *of the same type* (client or server) with the same public name that an
5491 error should be diagnosed.
5493 9. When Exim looked up a host name for an IP address, but failed to find the
5494 original IP address when looking up the host name (a safety check), it
5495 output the message "<ip address> does not match any IP for NULL", which was
5496 confusing, to say the least. The bug was that the host name should have
5497 appeared instead of "NULL".
5499 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
5500 user that is built into the binary, and the -C or -D options is used, root
5501 privilege is dropped before the configuration file is read. In addition,
5502 logging is switched to stderr instead of the normal log files. If the
5503 configuration then re-defines the Exim user, the unprivileged environment
5504 is probably not what is expected, so Exim logs a panic warning message (but
5507 However, if deliver_drop_privilege is set, the unprivileged state may well
5508 be exactly what is intended, so the warning has been cut out in that case,
5509 and Exim is allowed to try to write to its normal log files.
5515 1. smtp_return_error_details was not giving details for temporary sender
5516 or receiver verification errors.
5518 2. Diagnose a configuration error if two authenticators have the same public
5521 3. Exim used not to create the message log file for a message until the first
5522 delivery attempt. This could be confusing when incoming messages were held
5523 for policy or load reasons. The message log file is now created at the time
5524 the message is received, and an initial "Received" line is written to it.
5526 4. The automatically generated man page for command line options had a minor
5527 bug that caused no ill effects; however, a more serious problem was that
5528 the procedure for building the man page automatically didn't always
5529 operate. Consequently, release 4.20 contains an out-of-date version. This
5530 shouldn't happen again.
5532 5. When building Exim with embedded Perl support, the script that builds the
5533 Makefile was calling 'perl' to find its compile-time parameters, ignoring
5534 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
5536 6. The freeze_tell option was not being used for messages that were frozen on
5537 arrival, either by an ACL or by local_scan().
5539 7. Added the smtp_incomplete_transaction log selector.
5541 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
5542 was accepting AUTH without a new EHLO.
5544 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
5545 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
5548 10. Logging of TCP/IP connections (when configured) now happens in the main
5549 daemon process instead of the child process, so that the TCP/IP connection
5550 count is more accurate (but it can never be perfect).
5552 11. The use of "drop" in a nested ACL was not being handled correctly in the
5553 outer ACL. Now, if condition failure induced by the nested "drop" causes
5554 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
5555 or "require"), the connection is dropped.
5557 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
5558 that yield "discard" can now be used with an "accept" or a "discard" verb,
5559 but an error is generated for any others (because I can't see a useful way
5560 to define what should happen).
5562 13. When an ACL is read dynamically from a file (or anywhere else), the lines
5563 are now processed in the same way as lines in the Exim configuration file.
5564 In particular, continuation lines are supported.
5566 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
5568 15. Added -ti meaning -t -i.
5570 16. Check for letters, digits, hyphens, and dots in the names of dnslist
5571 domains, and warn by logging if others are found.
5573 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
5574 returned from GIFCONF when Exim is trying to find the list of interfaces on
5575 a host. The code in os.c has been modified to copy each ifreq to an aligned
5576 structure in all cases.
5578 Also, in some cases, the returned ifreq's were being copied to a 'struct
5579 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
5580 means the last couple of bytes of an IPv6 address could be chopped if the
5581 ifreq contained only a normal sockaddr (14 bytes storage).
5583 18. Named domain lists were not supported in the hosts_treat_as_local option.
5584 An entry such as +xxxx was not recognized, and was treated as a literal
5587 19. Ensure that header lines added by a DATA ACL are included in the reject log
5588 if the ACL subsequently rejects the message.
5590 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
5591 MD5 (which is deprecated).
5593 21. When testing a filter file using -bf, Exim was writing a message when it
5594 took the sender from a "From " line in the message, but it was not doing so
5595 when it took $return_path from a Return-Path: header line. It now does.
5597 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
5598 with a valid header line field name (a series of printing characters
5599 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
5601 23. Changed "disc" in the source to "disk" to conform to the documentation and
5602 the book and for uniformity.
5604 24. Ignore Sendmail's -Ooption=value command line item.
5606 25. When execve() failed while trying to run a command in a pipe transport,
5607 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
5608 could be confused with a return value of 69 from the command itself. This
5609 has been changed to 127, the value the shell returns if it is asked to run
5610 a non-existent command. The wording for the related log line suggests a
5611 non-existent command as the problem.
5613 26. If received_header_text expands to an empty string, do not add a Received:
5614 header line to the message. (Well, it adds a token one on the spool, but
5615 marks it "old" so that it doesn't get used or transmitted.)
5617 27. Installed eximstats 1.28 (addition of -nt option).
5619 28. There was no check for failure on the call to getsockname() in the daemon
5620 code. This can fail if there is a shortage of resources on the system, with
5621 ENOMEM, for example. A temporary error is now given on failure.
5623 29. Contrary to the C standard, it seems that in some environments, the
5624 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
5625 program. Exim now does this explicitly; it affects the formatting of
5626 timestamps using strftime().
5628 30. If exiqsumm was given junk data, it threw up some uninitialized variable
5629 complaints. I've now initialized all the variables, to avoid this.
5631 32. Header lines added by a system filter were not being "seen" during
5632 transport-time rewrites.
5634 33. The info_callback() function passed to OpenSSL is set up with type void
5635 (*)(SSL *, int, int), as described somewhere. However, when calling the
5636 function (actually a macro) that sets it up, the type void(*)() is
5637 expected. I've put in a cast to prevent warnings from picky compilers.
5639 34. If a DNS black list lookup found a CNAME record, but there were no A
5640 records associated with the domain it pointed at, Exim crashed.
5642 35. If a DNS black list lookup returned more than one A record, Exim ignored
5643 all but the first. It now scans all returned addresses if a particular IP
5644 value is being sought. In this situation, the contents of the
5645 $dnslist_value variable are a list of all the addresses, separated by a
5648 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
5649 to accept a host name and all its aliases if the forward lookup for any of
5650 them yielded the IP address of the incoming connection. Now it accepts only
5651 those names whose forward lookup yields the correct IP address. Any other
5652 names are discarded. This closes a loophole whereby a rogue DNS
5653 administrator could create reverse DNS records to break through a
5654 wildcarded host restriction in an ACL.
5656 37. If a user filter or a system filter that ran in a subprocess used any of
5657 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
5658 the wrong values were passed to the pipe command ($thisaddress had the
5659 value of $0, $0 had the value of $1, etc). This bug was introduced by
5660 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
5662 38. Improved the line breaking for long SMTP error messages from ACLs.
5663 Previously, if there was no break point between 40 and 75 characters, Exim
5664 left the rest of the message alone. Two changes have been made: (a) I've
5665 reduced the minimum length to 35 characters; (b) if it can't find a break
5666 point between 35 and 75 characters, it looks ahead and uses the first one
5667 that it finds. This may give the occasional overlong line, but at least the
5668 remaining text gets split now.
5670 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
5671 file descriptors might be low, and that setting 1000 would always raise it.
5672 It turns out that in some environments, the limit is already over 1000 and
5673 that lowering it causes trouble. So now Exim takes care not to decrease it.
5675 40. When delivering a message, the value of $return_path is set to $sender_
5676 address at the start of routing (routers may change the value). By an
5677 oversight, this default was not being set up when an address was tested by
5678 -bt or -bv, which affected the outcome if any router or filter referred to
5681 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
5682 log only when "message" or "log_message" are set. However, if one of the
5683 conditions was an address verification, or a call to a nested ACL, the
5684 messages generated by the underlying test were being passed through. This
5685 no longer happens. The underlying message is available in $acl_verify_
5686 message for both "message" and "log_message" expansions, so it can be
5687 passed through if needed.
5689 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
5690 new expansion $bh_ to give the encoded byte string without charset
5691 translation. Translation happens only if iconv() is available; HAVE_ICONV
5692 indicates this at build time. HEADERS_CHARSET gives the charset to
5693 translate to; headers_charset can change it in the configuration, and
5694 "headers charset" can change it in an individual filter file.
5696 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
5697 that creates RFC 2047 encoded "words" labels them as that charset instead
5698 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
5699 expansion operator; (ii) when Exim creates a From: line for a local
5700 message; (iii) when a header line is rewritten to include a "phrase" part.
5702 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
5703 buggy, causing it to skip the first lines of messages whose message ID
5704 ended in 'D'. This would not have bitten before Exim release 4.14, because
5705 message IDs were unlikely to end in 'D' before then. The effect was to have
5706 incorrect size information for certain domains.
5708 45. #include "config.h" was missing at the start of the crypt16.c module. This
5709 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
5712 46. If there was a timeout during a "random" callout check, Exim treated it as
5713 a failure of the random address, and carried on sending RSET and the real
5714 address. If the delay was just some slowness somewhere, the response to the
5715 original RCPT would be taken as a response to RSET and so on, causing
5716 mayhem of various kinds.
5718 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
5719 when I implemented it. It didn't allow for the fact that some option values
5720 may legitimatetly be negative (e.g. size_addition), and it didn't even do
5721 the right test for positive values.
5723 48. Domain names in DNS records are case-independent. Exim always looks them up
5724 in lower case. Some resolvers return domain names in exactly the case they
5725 appear in the zone file, that is, they may contain uppercase letters. Not
5726 all resolvers do this - some return always lower case. Exim was treating a
5727 change of case by a resolver as a change of domain, similar to a widening
5728 of a domain abbreviation. This triggered its re-routing code and so it was
5729 trying to route what was effectively the same domain again. This normally
5730 caused routing to fail (because the router wouldn't handle the domain
5731 twice). Now Exim checks for this case specially, and just changes the
5732 casing of the domain that it ultimately uses when it transmits the message
5735 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
5738 50. If a filter generated a file delivery with a non-absolute name (possible if
5739 no home directory exists for the router), the forbid_file option was not
5742 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
5743 the existing equality matching.
5745 52. Exim was using ints instead of ino_t variables in some places where it was
5746 dealing with inode numbers.
5748 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
5749 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
5750 called TMPDIR, and if it finds it is different, it changes its value.
5752 54. The smtp_printf() function is now made available to local_scan() so
5753 additional output lines can be written before returning. There is also an
5754 smtp_fflush() function to enable the detection of a dropped connection.
5755 The variables smtp_input and smtp_batched_input are exported to
5758 55. Changed the default runtime configuration: the message "Unknown user"
5759 has been removed from the ACL, and instead placed on the localuser router,
5760 using the cannot_route_message feature. This means that any verification
5761 failures that generate their own messages won't get overridden. Similarly,
5762 the "Unrouteable address" message that was in the ACL for unverifiable
5763 relay addresses has also been removed.
5765 56. Added hosts_avoid_esmtp to the smtp transport.
5767 57. The exicyclog script was not checking for the esoteric option
5768 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
5769 will work only if exicyclog is run under the appropriate euid.
5771 58. Following a discussion on the list, the rules by which Exim recognises line
5772 endings on incoming messages have been changed. The -dropcr and drop_cr
5773 options are now no-ops, retained only for backwards compatibility. The
5774 following line terminators are recognized: LF CRLF CR. However, special
5775 processing applies to CR:
5777 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
5778 nor a local message in the state where . is a terminator.
5780 (ii) If a bare CR is encountered in a header line, an extra space is added
5781 after the line terminator so as not to end the header. The reasoning
5782 behind this is that bare CRs in header lines are most likely either
5783 to be mistakes, or people trying to play silly games.
5785 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
5786 was being incorrectly given as 18 bytes larger than it should have been.
5787 This is a VOB (very old bug).
5789 60. This may never have affected anything current, but just in case it has:
5790 When the local host is found other than at the start of a list of hosts,
5791 the local host, those with the same MX, and any that follow, are discarded.
5792 When the list in question was part of a longer list of hosts, the following
5793 hosts (not currently being processed) were also being discarded. This no
5794 longer happens. I'm not sure if this situation could ever has previously
5797 61. Added the "/MX" feature to lists of hosts in the manualroute and query
5800 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
5801 header. This is something that is recommended in a new Internet Draft, and
5802 is something that is documented as being done by Sendmail. There are two
5803 possible values. For messages generated by the autoreply transport, Exim
5806 Auto-Submitted: auto-replied
5808 whereas for all other generated messages (e.g. bounces) it adds
5810 Auto-Submitted: auto-generated
5812 63. The "personal" condition in filters now includes a test for the
5813 Auto-Submitted: header. If it contains the string "auto-" the message it
5814 not considered personal.
5816 64. Added rcpt_include_affixes as a generic transport option.
5818 65. Added queue_only_override (default true).
5820 66. Added the syslog_duplication option.
5822 67. If what should have been the first header line of a message consisted of
5823 a space followed by a colon, Exim was mis-interpreting it as a header line.
5824 It isn't of course - it is syntactically invalid and should therefore be
5825 treated as the start of the message body. The misbehaviour could have
5826 caused a number of strange effects, including loss of data in subsequent
5827 header lines, and spool format errors.
5829 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
5830 client host had authenticated. This control can now be exercised by an ACL
5831 for more flexibility.
5833 69. By default, callouts do not happen when testing with -bh. There is now a
5834 variant, -bhc, which does actually run the callout code, including
5835 consulting and updating the callout cache.
5837 70. Added support for saslauthd authentication, courtesy of Alexander
5840 71. If statvfs() failed on the spool or log directories while checking their
5841 size for availability, Exim confusingly gave the error "space shortage".
5842 Furthermore, in debugging mode it crashed with a floating point exception.
5843 These checks are done if check_{spool,log}_{space,inodes} are set, and when
5844 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
5845 serious problem, Exim now writes to the main and panic logs when this
5846 happens, with details of the failure. It then refuses to accept the
5847 incoming message, giving the message "spool directory problem" or "log
5848 directory problem" with a 421 code for SMTP messages.
5850 72. When Exim is about to re-exec itself, it ensures that the file descriptors
5851 0, 1, and 2 exist, because some OS complain for execs without them (see
5852 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
5853 descriptors. However, the code omitted to check that the open succeeded,
5854 causing mysterious errors if for some reason the permissions on /dev/null
5855 got screwed. Now Exim writes a message to the main and panic logs, and
5856 bombs out if it can't open /dev/null.
5858 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
5859 interact so that it is all more flexible. It is supposed to remain
5860 backwards compatible. Also added extra_local_interfaces.
5862 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
5863 to bomb out with an assertion failure - to the client this appears as a
5864 connection drop. This problem occurs in the part of the code that was taken
5865 from the Samba project. Fortunately, the assertion is in a very simple
5866 function, so I have fixed this by reproducing the function inline in the
5867 one place where it is called, and arranging for authentication to fail
5868 instead of killing the process with assert().
5870 75. The SPA client code was not working when the server requested OEM rather
5871 than Unicode encoding.
5873 76. Added code to make require_files with a specific uid setting more usable in
5874 the case where statting the file as root fails - usually a non-root-mounted
5875 NFS file system. When this happens and the failure is EACCES, Exim now
5876 forks a subprocess and does the per-uid checking as the relevant uid.
5878 77. Added process_log_path.
5880 78. If log_file_path was not explicitly set, a setting of check_log_space or
5881 check_log_inodes was ignored.
5883 79. If a space check for the spool or log partitions fails, the incident is now
5884 logged. Of course, in the latter case the data may get lost...
5886 80. Added the %p formatting code to string_format() so that it can be used to
5887 print addresses in debug_print(). Adjusted all the address printing in the
5888 debugging in store.c to use %p rather than %d.
5890 81. There was a concern that a line of code in smtp_in.c could overflow a
5891 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
5892 initially expressed, the concern was not well-founded, because trailing
5893 spaces are removed early. However, if the trailing spaces were followed by
5894 a NULL, they did not get removed, so the overflow was possible. Two fixes
5897 (a) I re-wrote the offending code in a cleaner fashion.
5898 (b) If an incoming SMTP command contains a NULL character, it is rejected
5901 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
5902 runs initgroups(), so that if the Exim user is in any additional groups,
5903 they will be used during message reception.
5909 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / jgh / exim.git / blob