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 PP/21 Fix eximon continuous updating with timestamped log-files.
201 Broken in a format-string cleanup in 4.80, missed when I repaired the
202 other false fix of the same issue.
203 Report and fix from Heiko Schlichting.
206 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
207 Report from Prashanth Katuri.
209 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
210 It's SecureTransport, so affects any MacOS clients which use the
211 system-integrated TLS libraries, including email clients.
213 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
214 using a MIME ACL for non-SMTP local injection.
215 Report and assistance in diagnosis by Warren Baker.
217 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
219 JH/16 Fix comparisons for 64b. Bugzilla 1385.
221 TL/09 Add expansion variable $authenticated_fail_id to keep track of
222 last id that failed so it may be referenced in subsequent ACL's.
224 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
227 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
228 ldap library initialization, allowing self-signed CA's to be
229 used. Also properly sets require_cert option later in code by
230 using NULL (global ldap config) instead of ldap handle (per
231 session). Bug diagnosis and testing by alxgomz.
237 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
239 This, or similar/improved, will also be change PP/11 of 4.82.
245 PP/01 Handle short writes when writing local log-files.
246 In practice, only affects FreeBSD (8 onwards).
247 Bugzilla 1053, with thanks to Dmitry Isaikin.
249 NM/01 Bugzilla 949 - Documentation tweak
251 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
254 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
256 PP/02 Implemented gsasl authenticator.
258 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
260 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
261 `pkg-config foo` for cflags/libs.
263 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
264 with rest of GSASL and with heimdal_gssapi.
266 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
267 `pkg-config foo` for cflags/libs for the TLS implementation.
269 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
270 properties get this fed in as external SSF. A number of robustness
271 and debugging improvements to the cyrus_sasl authenticator.
273 PP/08 cyrus_sasl server now expands the server_realm option.
275 PP/09 Bugzilla 1214 - Log authentication information in reject log.
276 Patch by Jeremy Harris.
278 PP/10 Added dbmjz lookup type.
280 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
282 PP/12 MAIL args handles TAB as well as SP, for better interop with
283 non-compliant senders.
284 Analysis and variant patch by Todd Lyons.
286 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
287 Bug report from Lars Müller <lars@samba.org> (via SUSE),
288 Patch from Dirk Mueller <dmueller@suse.com>
290 PP/13 tls_peerdn now print-escaped for spool files.
291 Observed some $tls_peerdn in wild which contained \n, which resulted
292 in spool file corruption.
294 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
295 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
296 or write after TLS renegotiation, which otherwise led to messages
299 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
300 as a tracking header (ie: a signed header comes before the signature).
301 Patch from Wolfgang Breyha.
303 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
304 comma-sep list; embedded commas doubled.
306 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
308 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
310 Report and patch from Dmitry Banschikov.
312 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
313 Removed SSL_clear() after SSL_new() which led to protocol negotiation
314 failures. We appear to now support TLS1.1+ with Exim.
316 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
317 lets Exim select keys and certificates based upon TLS SNI from client.
318 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
319 before an outbound SMTP session. New log_selector, +tls_sni.
321 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
322 NULL dereference. Report and patch from Alun Jones.
324 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
325 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
326 Not seeing resolver debug output on NetBSD, but suspect this is a
327 resolver implementation change.
329 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
330 Left warnings. Added "eximon gdb" invocation mode.
332 PP/21 Defaulting "accept_8bitmime" to true, not false.
334 PP/22 Added -bw for inetd wait mode support.
336 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
337 locate the relevant includes and libraries. Made this the default.
339 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
340 Bugzilla 1246, report and most of solution from Tomasz Kusy.
342 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
343 This may cause build issues on older platforms.
345 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
346 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
347 gnutls_require_mac & gnutls_require_protocols (no longer supported).
348 Added SNI support via GnuTLS too.
349 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
351 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
353 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
354 Applied second patch from Janne, implementing suggestion to default
355 multiple-strings-in-record handling to match SPF spec.
357 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
359 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
360 read-only, out of scope).
361 Patch from Wolfgang Breyha, report from Stuart Northfield.
363 PP/29 Fix three issues highlighted by clang analyser static analysis.
364 Only crash-plausible issue would require the Cambridge-specific
365 iplookup router and a misconfiguration.
366 Report from Marcin Mirosław.
368 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
370 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
371 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
372 As part of this, removing so much warning spew let me fix some minor
373 real issues in debug logging.
375 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
376 assignment on my part. Fixed.
378 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
379 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
380 Janne Snabb (who went above and beyond: thank you).
382 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
383 string otherwise requires a connection and a bunch more work and it's
384 relatively easy to get wrong. Should also expose TLS library linkage
387 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
388 64-bit ${eval} (JH/03).
390 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
391 GNU libc to support some of the 64-bit stuff, should not lead to
392 conflicts. Defined before os.h is pulled in, so if a given platform
393 needs to override this, it can.
395 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
396 protection layer was required, which is not implemented.
397 Bugzilla 1254, patch from Wolfgang Breyha.
399 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
400 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
401 tls_dhparam take prime identifiers. Also unbreak combination of
402 OpenSSL+DH_params+TLSSNI.
404 PP/39 Disable SSLv2 by default in OpenSSL support.
410 PP/01 Solaris build fix for Oracle's LDAP libraries.
411 Bugzilla 1109, patch from Stephen Usher.
413 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
415 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
418 TF/02 Fix a couple more cases where we did not log the error message
419 when unlink() failed. See also change 4.74-TF/03.
421 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
422 lock up or crash if it happened to be inside a call to libc when it
423 got a SIGUSR1 from exiwhat.
425 The SIGUSR1 handler appends the current process status to the process
426 log which is later printed by exiwhat. It used to use the general
427 purpose logging code to do this, but several functions it calls are
428 not safe for signals.
430 The new output code in the SIGUSR1 handler is specific to the process
431 log, and simple enough that it's easy to inspect for signal safety.
432 Removing some special cases also simplifies the general logging code.
433 Removing the spurious timestamps from the process log simplifies
436 TF/04 Improved ratelimit ACL condition.
438 The /noupdate option has been deprecated in favour of /readonly which
439 has clearer semantics. The /leaky, /strict, and /readonly update modes
440 are mutually exclusive. The update mode is no longer included in the
441 database key; it just determines when the database is updated. (This
442 means that when you upgrde Exim will forget old rate measurements.)
444 Exim now checks that the per_* options are used with an update mode that
445 makes sense for the current ACL. For example, when Exim is processing a
446 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
447 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
448 must specify per_mail/readonly. If you omit the update mode it defaults to
449 /leaky where that makes sense (as before) or /readonly where required.
451 The /noupdate option is now undocumented but still supported for
452 backwards compatibility. It is equivalent to /readonly except that in
453 ACLs where /readonly is required you may specify /leaky/noupdate or
454 /strict/noupdate which are treated the same as /readonly.
456 A useful new feature is the /count= option. This is a generalization
457 of the per_byte option, so that you can measure the throughput of other
458 aggregate values. For example, the per_byte option is now equivalent
459 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
461 The per_rcpt option has been generalized using the /count= mechanism
462 (though it's more complicated than the per_byte equivalence). When it is
463 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
464 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
465 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
466 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
467 Note that using per_rcpt with a non-readonly update mode in more than
468 one ACL will cause the recipients to be double-counted. (The per_mail
469 and per_byte options don't have this problem.)
471 The handling of very low rates has changed slightly. If the computed rate
472 is less than the event's count (usually one) then this event is the first
473 after a long gap. In this case the rate is set to the same as this event's
474 count, so that the first message of a spam run is counted properly.
476 The major new feature is a mechanism for counting the rate of unique
477 events. The new per_addr option counts the number of different
478 recipients that someone has sent messages to in the last time period. It
479 behaves like per_rcpt if all the recipient addresses are different, but
480 duplicate recipient addresses do not increase the measured rate. Like
481 the /count= option this is a general mechanism, so the per_addr option
482 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
483 example, measure the rate that a client uses different sender addresses
484 with the options per_mail/unique=$sender_address. There are further
485 details in the main documentation.
487 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
489 TF/06 Removed a few PCRE remnants.
491 TF/07 Automatically extract Exim's version number from tags in the git
492 repository when doing development or release builds.
494 PP/02 Raise smtp_cmd_buffer_size to 16kB.
495 Bugzilla 879. Patch from Paul Fisher.
497 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
498 Heavily based on revision 40f9a89a from Simon Arlott's tree.
501 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
503 PP/05 Variable $av_failed, true if the AV scanner deferred.
504 Bugzilla 1078. Patch from John Horne.
506 PP/06 Stop make process more reliably on build failure.
507 Bugzilla 1087. Patch from Heiko Schlittermann.
509 PP/07 Make maildir_use_size_file an _expandable_ boolean.
510 Bugzilla 1089. Patch from Heiko Schlittermann.
512 PP/08 Handle ${run} returning more data than OS pipe buffer size.
513 Bugzilla 1131. Patch from Holger Weiß.
515 PP/09 Handle IPv6 addresses with SPF.
516 Bugzilla 860. Patch from Wolfgang Breyha.
518 PP/10 GnuTLS: support TLS 1.2 & 1.1.
520 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
523 PP/11 match_* no longer expand right-hand-side by default.
524 New compile-time build option, EXPAND_LISTMATCH_RHS.
525 New expansion conditions, "inlist", "inlisti".
527 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
529 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
531 PP/14 fix log_write() format string regression from TF/03.
532 Bugzilla 1152. Patch from Dmitry Isaikin.
538 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
540 PP/02 Harmonised TLS library version reporting; only show if debugging.
541 Layout now matches that introduced for other libraries in 4.74 PP/03.
543 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
545 PP/04 New "dns_use_edns0" global option.
547 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
550 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
551 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
553 TK/01 Updated PolarSSL code to 0.14.2.
554 Bugzilla 1097. Patch from Andreas Metzler.
556 PP/07 Catch divide-by-zero in ${eval:...}.
559 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
562 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
563 format-string attack -- SECURITY: remote arbitrary code execution.
565 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
566 time unintentionally subject to list matching rules, letting the header
567 cause arbitrary Exim lookups (of items which can occur in lists, *not*
568 arbitrary string expansion). This allowed for information disclosure.
570 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
571 INT_MIN/-1 -- value coerced to INT_MAX.
577 NM/01 Workround for PCRE version dependency in version reporting
580 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
581 This fixes portability to compilers other than gcc, notably
582 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
584 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
585 makefiles for portability to HP-UX and POSIX correctness.
587 PP/01 Permit LOOKUP_foo enabling on the make command-line.
588 Also via indented variable definition in the Makefile.
589 (Debugging by Oliver Heesakkers).
591 PP/02 Restore caching of spamd results with expanded spamd_address.
592 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
594 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
595 Improves build reliability. Fix from: Frank Elsner
597 NM/02 Fix wide character breakage in the rfc2047 coding
598 Fixes bug 1064. Patch from Andrey N. Oktyabrski
600 NM/03 Allow underscore in dnslist lookups
601 Fixes bug 1026. Patch from Graeme Fowler
603 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
604 Code patches from Adam Ciarcinski of NetBSD.
606 NM/04 Fixed exiqgrep to cope with mailq missing size issue
609 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
610 is logged, to avoid truncation. Patch from John Horne.
612 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
613 Patch from Jakob Hirsch.
615 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
616 SQL string expansion failure details.
617 Patch from Andrey Oktyabrski.
619 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
620 Patch from Simon Arlott.
622 PP/09 New lookups functionality failed to compile on old gcc which rejects
623 extern declarations in function scope.
624 Patch from Oliver Fleischmann
626 PP/10 Use sig_atomic_t for flags set from signal handlers.
627 Check getgroups() return and improve debugging.
628 Fixed developed for diagnosis in bug 927 (which turned out to be
631 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
632 Patch from Mark Zealey.
634 PP/12 Bugzilla 1056: Improved spamd server selection.
635 Patch from Mark Zealey.
637 PP/13 Bugzilla 1086: Deal with maildir quota file races.
638 Based on patch from Heiko Schlittermann.
640 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
641 Patch from Uwe Doering, sign-off by Michael Haardt.
643 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
644 variable declaration deep within a block. Bug and patch from
647 PP/15 lookups-Makefile IRIX compatibilty coercion.
649 PP/16 Make DISABLE_DKIM build knob functional.
651 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
652 Patch by Simon Arlott
654 TF/03 Fix valgrind.h portability to C89 compilers that do not support
655 variable argument macros. Our copy now differs from upstream.
661 TF/01 Failure to get a lock on a hints database can have serious
662 consequences so log it to the panic log.
664 TF/02 Log LMTP confirmation messages in the same way as SMTP,
665 controlled using the smtp_confirmation log selector.
667 TF/03 Include the error message when we fail to unlink a spool file.
669 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
670 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
671 for maintaining out-of-tree patches for some time.
673 PP/01 Bugzilla 139: Documentation and portability issues.
674 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
675 Handle per-OS dynamic-module compilation flags.
677 PP/02 Let /dev/null have normal permissions.
678 The 4.73 fixes were a little too stringent and complained about the
679 permissions on /dev/null. Exempt it from some checks.
680 Reported by Andreas M. Kirchwitz.
682 PP/03 Report version information for many libraries, including
683 Exim version information for dynamically loaded libraries. Created
684 version.h, now support a version extension string for distributors
685 who patch heavily. Dynamic module ABI change.
687 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
688 privilege escalation vulnerability whereby the Exim run-time user
689 can cause root to append content of the attacker's choosing to
692 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
695 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
696 If dropping privileges for untrusted macros, we disabled normal logging
697 on the basis that it would fail; for the Exim run-time user, this is not
698 the case, and it resulted in successful deliveries going unlogged.
699 Fixed. Reported by Andreas Metzler.
705 PP/01 Date: & Message-Id: revert to normally being appended to a message,
706 only prepend for the Resent-* case. Fixes regression introduced in
707 Exim 4.70 by NM/22 for Bugzilla 607.
709 PP/02 Include check_rfc2047_length in configure.default because we're seeing
710 increasing numbers of administrators be bitten by this.
712 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
714 PP/03 Bugzilla 994: added openssl_options main configuration option.
716 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
718 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
720 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
722 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
723 without a peer certificate, leading to a segfault because of an
724 assumption that peers always have certificates. Be a little more
725 paranoid. Problem reported by Martin Tscholak.
727 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
728 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
729 NB: ClamAV planning to remove STREAM in "middle of 2010".
730 CL also introduces -bmalware, various -d+acl logging additions and
731 more caution in buffer sizes.
733 PP/09 Implemented reverse_ip expansion operator.
735 PP/10 Bugzilla 937: provide a "debug" ACL control.
737 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
739 PP/12 Bugzilla 973: Implement --version.
741 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
743 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
745 PP/15 Bugzilla 816: support multiple condition rules on Routers.
747 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
748 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
749 ignore trailing whitespace.
751 JJ/02 prevent non-panic DKIM error from being sent to paniclog
753 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
756 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
757 Notification from Dr Andrew Aitchison.
759 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
760 ExtendedDetectionInfo response format.
761 Notification from John Horne.
763 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
766 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
767 XSL and documented dependency on system catalogs, with examples of how
770 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
773 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
774 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
775 configuration file which is writeable by the Exim user or group.
777 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
778 of configuration files to cover files specified with the -C option if
779 they are going to be used with root privileges, not just the default
782 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
783 option (effectively making it always true).
785 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
786 files to be used while preserving root privileges.
788 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
789 that rogue child processes cannot use them.
791 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
792 run-time user, instead of root.
794 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
795 Exim run-time user without dropping privileges.
797 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
798 result string, instead of calling string_vformat() twice with the same
801 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
802 for other users. Others should always drop root privileges if they use
803 -C on the command line, even for a whitelisted configure file.
805 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
807 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
813 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
814 $data_path, and $header_path variables; fixed documentation bugs and
817 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
818 exipick to access non-standard spools, including the "frozen" queue
821 NM/01 Bugzilla 965: Support mysql stored procedures.
822 Patch from Alain Williams
824 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
826 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
827 Patch from Andreas Metzler
829 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
830 Patch from Kirill Miazine
832 NM/05 Bugzilla 671: Added umask to procmail example.
834 JJ/03 installed exipick 20100323.0, fixing doc bug
836 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
837 directory. Notification and patch from Dan Rosenberg.
839 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
841 TK/02 Improve log output when DKIM signing operation fails.
843 MH/01 Treat the transport option dkim_domain as a colon separated
844 list, not as a single string, and sign the message with each element,
845 omitting multiple occurences of the same signer.
847 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
848 Bugzilla 985, 986. Patch by Simon Arlott
850 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
851 Patch by Simon Arlott
853 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
854 MBX locking. Notification from Dan Rosenberg.
860 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
862 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
864 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
866 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
868 NM/04 Bugzilla 915: Flag broken perl installation during build.
874 TK/01 Added patch by Johannes Berg that expands the main option
875 "spamd_address" if it starts with a dollar sign.
877 TK/02 Write list of recipients to X-Envelope-Sender header when building
878 the mbox-format spool file for content scanning (suggested by Jakob
881 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
882 (http://www.dcc-servers.net/) support via dccifd. Activated by
883 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
885 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
886 by Mark Daniel Reidel <mr@df.eu>.
888 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
889 When building exim an external PCRE library is now needed -
890 PCRE is a system library on the majority of modern systems.
891 See entry on PCRE_LIBS in EDITME file.
893 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
894 conversation. Added nologin parameter to request.
895 Patch contributed by Kirill Miazine.
897 TF/01 Do not log submission mode rewrites if they do not change the address.
899 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
901 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
902 log files in place. Contributed by Roberto Lima.
904 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
906 TF/03 Bugzilla 615: When checking the local_parts router precondition
907 after a local_part_suffix or local_part_prefix option, Exim now
908 does not use the address's named list lookup cache, since this
909 contains cached lookups for the whole local part.
911 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
912 Robert Millan. Documentation is in experimental-spec.txt.
914 TF/04 Bugzilla 668: Fix parallel build (make -j).
916 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
918 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
919 Patch provided by Jan Srzednicki.
921 TF/05 Leading white space used to be stripped from $spam_report which
922 wrecked the formatting. Now it is preserved.
924 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
925 that they are available at delivery time.
927 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
929 TF/08 TLS error reporting now respects the incoming_interface and
930 incoming_port log selectors.
932 TF/09 Produce a more useful error message if an SMTP transport's hosts
933 setting expands to an empty string.
935 NM/06 Bugzilla 744: EXPN did not work under TLS.
936 Patch provided by Phil Pennock.
938 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
939 Patch provided by Richard Godbee.
941 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
942 acl_smtp_notquit, added index entry.
944 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
945 Patch provided by Eugene Bujak.
947 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
948 accept(). Patch provided by Maxim Dounin.
950 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
951 Patch provided by Phil Pennock.
953 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
955 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
956 Patch provided by Brad "anomie" Jorsch.
958 NM/14 Bugzilla 622: Added timeout setting to transport filter.
959 Patch provided by Dean Brooks.
961 TK/05 Add native DKIM support (does not depend on external libraries).
963 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
964 Patch provided by Graeme Fowler.
966 NM/16 Bugzilla 851: Documentation example syntax fix.
968 NM/17 Changed NOTICE file to remove references to embedded PCRE.
970 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
973 NM/19 Bugzilla 745: TLS version reporting.
974 Patch provided by Phil Pennock.
976 NM/20 Bugzilla 167: bool: condition support.
977 Patch provided by Phil Pennock.
979 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
980 clients. Patch provided by Phil Pennock.
982 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
983 Patch provided by Brad "anomie" Jorsch.
985 NM/23 Bugzilla 687: Fix misparses in eximstats.
986 Patch provided by Heiko Schlittermann.
988 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
989 Patch provided by Heiko Schlittermann.
991 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
992 plus update to original patch.
994 NM/26 Bugzilla 799: Documentation correction for ratelimit.
996 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
997 Patch provided by David Brownlee.
999 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1001 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1003 NM/30 Bugzilla 888: TLS documentation bugfixes.
1005 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1007 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1008 Unlike the original bugzilla I have changed all shell scripts in src tree.
1010 NM/33 Bugzilla 898: Transport filter timeout fix.
1011 Patch by Todd Rinaldo.
1013 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1014 Patch by Serge Demonchaux.
1016 NM/35 Bugzilla 39: Base64 decode bug fixes.
1017 Patch by Jakob Hirsch.
1019 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1021 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1023 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1025 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1031 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1032 ALT-N's libdkim that I have put here:
1033 http://duncanthrax.net/exim-experimental/
1035 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1036 were called 'true' and it seems that C99 defines that as a reserved
1037 keyword to be used with 'bool' variable types. That means you could
1038 not include C99-style headers which use bools without triggering
1039 build errors in sieve.c.
1041 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1042 as mailq or other aliases. Changed the --help handling significantly
1043 to do whats expected. exim_usage() emits usage/help information.
1045 SC/01 Added the -bylocaldomain option to eximstats.
1047 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1049 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1051 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1057 PH/01 Another patch from the Sieve maintainer.
1059 PH/02 When an IPv6 address is converted to a string for single-key lookup
1060 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1061 dots are used instead of colons so that keys in lsearch files need not
1062 contain colons. This was done some time before quoting was made available
1063 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1064 (notated using the quote facility) so as to distinguish them from IPv4
1065 keys. This meant that lookups for IP addresses in host lists did not work
1066 for iplsearch lookups.
1068 This has been fixed by arranging for IPv6 addresses to be expressed with
1069 colons if the lookup type is iplsearch. This is not incompatible, because
1070 previously such lookups could never work.
1072 The situation is now rather anomolous, since one *can* have colons in
1073 ordinary lsearch keys. However, making the change in all cases is
1074 incompatible and would probably break a number of configurations.
1076 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1079 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1080 conversion specification without a maximum field width, thereby enabling
1081 a rogue spamd server to cause a buffer overflow. While nobody in their
1082 right mind would setup Exim to query an untrusted spamd server, an
1083 attacker that gains access to a server running spamd could potentially
1084 exploit this vulnerability to run arbitrary code as the Exim user.
1086 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1087 $primary_hostname instead of what libspf2 thinks the hosts name is.
1089 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1090 a directory entry by the name of the lookup key. Previously, if a
1091 symlink pointed to a non-existing file or a file in a directory that
1092 Exim lacked permissions to read, a lookup for a key matching that
1093 symlink would fail. Now it is enough that a matching directory entry
1094 exists, symlink or not. (Bugzilla 503.)
1096 PH/03 The body_linecount and body_zerocount variables are now exported in the
1099 PH/04 Added the $dnslist_matched variable.
1101 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1102 This means they are set thereafter only if the connection becomes
1105 PH/06 Added the client_condition to authenticators so that some can be skipped
1106 by clients under certain conditions.
1108 PH/07 The error message for a badly-placed control=no_multiline_responses left
1109 "_responses" off the end of the name.
1111 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1113 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1114 (without spaces) instead of just copying the configuration text.
1116 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1118 PH/11 Added $max_received_linelength.
1120 PH/12 Added +ignore_defer and +include_defer to host lists.
1122 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1123 way in which PCRE > 7.0 is built.
1125 PH/14 Implemented queue_only_load_latch.
1127 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1128 MAIL command. The effect was to mangle the value on 64-bit systems.
1130 PH/16 Another patch from the Sieve maintainer.
1132 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1134 PH/18 If a system quota error occurred while trying to create the file for
1135 a maildir delivery, the message "Mailbox is full" was not appended to the
1136 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1137 only to a quota excession during the actual writing of the file.
1139 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1140 characters?) which causes problems in log lines. The DN values are now
1141 passed through string_printing() before being added to log lines.
1143 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1144 and InterBase are left for another time.)
1146 PH/21 Added message_body_newlines option.
1148 PH/22 Guard against possible overflow in moan_check_errorcopy().
1150 PH/23 POSIX allows open() to be a macro; guard against that.
1152 PH/24 If the recipient of an error message contained an @ in the local part
1153 (suitably quoted, of course), incorrect values were put in $domain and
1154 $local_part during the evaluation of errors_copy.
1160 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1161 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1164 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1165 issue a MAIL command.
1167 PH/02 In an ACL statement such as
1169 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1171 if a client was not listed at all, or was listed with a value other than
1172 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1173 the condition was not true (as it should be), so access was not denied.
1174 The bug was that the ! inversion was incorrectly passed on to the second
1175 item. This has been fixed.
1177 PH/03 Added additional dnslists conditions == and =& which are different from
1178 = and & when the dns lookup returns more than one IP address.
1180 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1181 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1183 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1184 FSYNC, which compiles an option called disable_fsync that allows for
1185 bypassing fsync(). The documentation is heavily laced with warnings.
1187 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1189 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1190 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1191 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1192 including adding "make clean"; (3) Added -fPIC when compiling the test
1193 dynamically loaded module, to get rid of a warning.
1195 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1196 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1197 = 0s. The bug is otherwise harmless.
1199 PH/07 There was a bug in the dovecot authenticator such that the value of
1200 $auth1 could be overwritten, and so not correctly preserved, after a
1201 successful authentication. This usually meant that the value preserved by
1202 the server_setid option was incorrect.
1204 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1206 PH/09 Installed PCRE release 7.0.
1208 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1209 run for batched SMTP input. It is now run at the start of every message
1210 in the batch. While fixing this I discovered that the process information
1211 (output by running exiwhat) was not always getting set for -bs and -bS
1212 input. This is fixed, and it now also says "batched" for BSMTP.
1214 PH/11 Added control=no_pipelining.
1216 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1217 patch, slightly modified), and move the expansion of helo_data till after
1218 the connection is made in the smtp transport (so it can use these
1221 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1223 PH/14 Added log_selector = +pid.
1225 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1227 PH/16 Add ${if forany and ${if forall.
1229 PH/17 Added dsn_from option to vary the From: line in DSNs.
1231 PH/18 Flush SMTP output before performing a callout, unless control =
1232 no_callout_flush is set.
1234 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1235 was true (the default) a successful delivery failed to delete the retry
1236 item, thus causing premature timeout of the address. The bug is now
1239 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1241 PH/21 Long custom messages for fakedefer and fakereject are now split up
1242 into multiline reponses in the same way that messages for "deny" and
1243 other ACL rejections are.
1245 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1246 with slight modification.
1248 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1249 draft, changing the syntax and factoring some duplicate code".
1251 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1252 for deliveries of the second and subsequent messages over the same SMTP
1255 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1256 ${reduce, with only minor "tidies".
1258 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1260 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1261 expansion side effects.
1263 PH/27 When a message times out after an over-quota error from an Exim-imposed
1264 quota, the bounce message says "mailbox is full". This message was not
1265 being given when it was a system quota that was exceeded. It now should
1268 MH/03 Made $recipients available in local_scan(). local_scan() already has
1269 better access to the recipient list through recipients_list[], but
1270 $recipients can be useful in postmaster-provided expansion strings.
1272 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1273 in the case of a MAIL command with additional options following the
1274 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1275 were accidentally chopped off.
1277 PH/29 SMTP synchronization checks are implemented when a command is read -
1278 there is a check that no more input is waiting when there shouldn't be
1279 any. However, for some commands, a delay in an ACL can mean that it is
1280 some time before the response is written. In this time, more input might
1281 arrive, invalidly. So now there are extra checks after an ACL has run for
1282 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1283 pipelining has not been advertised.
1285 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1287 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1288 returned in the SMTP response when smtp_return_error_details was false.
1289 This has been fixed.
1291 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1292 instead of the C I/O that was originally supplied, because problems were
1293 reported on Solaris.
1295 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1296 Exim which did not show up earlier: it was assuming that a call to
1297 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1298 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1299 was a macro that became an assignment, so it seemed to work. This has
1300 changed to a proper function call with a void return, hence the compile
1301 error. Exim's code has been fixed.
1303 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1306 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1308 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1310 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1311 because it turns out to be too big or there is a timeout) there is no
1312 "Completed" line in the log. When some messages of this type were
1313 selected by exigrep, they were listed as "not completed". Others were
1314 picked up by some special patterns. I have improved the selection
1315 criteria to be more general.
1317 PH/38 The host_find_failed option in the manualroute router can now be set
1318 to "ignore", to completely ignore a host whose IP address cannot be
1319 found. If all hosts are ignored, the behaviour is controlled by the new
1320 host_all_ignored option.
1322 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1323 homing or because of multiple MX records with /mx) generated more than
1324 one IP address, and the following item turned out to be the local host,
1325 all the secondary addresses of the first item were incorrectly removed
1326 from the list, along with the local host and any following hosts (which
1327 is what is supposed to happen).
1329 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1330 whoever called Exim into the -H file. In the case of the daemon it was
1331 behaving confusingly. When first started, it used values for whoever
1332 started the daemon, but after a SIGHUP it used the Exim user (because it
1333 calls itself on a restart). I have changed the code so that it now always
1336 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1337 message are rejected with the same error (e.g. no authentication or bad
1338 sender address), and a DATA command is nevertheless sent (as can happen
1339 with PIPELINING or a stupid MUA), the error message that was given to the
1340 RCPT commands is included in the rejection of the DATA command. This is
1341 intended to be helpful for MUAs that show only the final error to their
1344 PH/42 Another patch from the Sieve maintainer.
1346 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1347 Eximstats - Fixed some broken HTML links and added missing column headers
1349 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1350 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1352 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1354 PH/43 Yet another patch from the Sieve maintainer.
1356 PH/44 I found a way to check for a TCP/IP connection going away before sending
1357 the response to the final '.' that terminates a message, but only in the
1358 case where the client has not sent further data following the '.'
1359 (unfortunately, this is allowed). However, in many cases there won't be
1360 any further data because there won't be any more messages to send. A call
1361 to select() can be used: if it shows that the input is "ready", there is
1362 either input waiting, or the socket has been closed. An attempt to read
1363 the next input character can distinguish the two cases. Previously, Exim
1364 would have sent an OK response which the client would never have see.
1365 This could lead to message repetition. This fix should cure that, at
1366 least in a lot of common cases.
1368 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1369 advertised in response to EHLO.
1375 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1376 fixed by 4.65/MH/01 (is this a record?) are fixed:
1378 (i) An empty string was always treated as zero by the numeric comparison
1379 operators. This behaviour has been restored.
1381 (ii) It is documented that the numeric comparison operators always treat
1382 their arguments as decimal numbers. This was broken in that numbers
1383 starting with 0 were being interpreted as octal.
1385 While fixing these problems I realized that there was another issue that
1386 hadn't been noticed. Values of message_size_limit (both the global option
1387 and the transport option) were treated as octal if they started with 0.
1388 The documentation was vague. These values are now always treated as
1389 decimal, and I will make that clear in the documentation.
1395 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1396 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1399 MH/01 Don't check that the operands of numeric comparison operators are
1400 integers when their expansion is in "skipping" mode (fixes bug
1401 introduced by 4.64-PH/07).
1403 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1404 child addresses, Exim now panics and dies. Previously, because the count
1405 is held in a short int, deliveries were likely to be lost. As such a
1406 large number of recipients for a single message is ridiculous
1407 (performance will be very, very poor), I have chosen to impose a limit
1408 rather than extend the field.
1414 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1415 leftover -K file (the existence of which was triggered by #402).
1416 While we were at it, introduced process PID as part of the -K
1417 filename. This should rule out race conditions when creating
1420 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1421 processing considerably. Previous code took too long for large mails,
1422 triggering a timeout which in turn triggers #401.
1424 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1425 in the DK code in transports.c. sendfile() is not really portable,
1426 hence the _LINUX specificness.
1428 TF/01 In the add_headers option to the mail command in an Exim filter,
1429 there was a bug that Exim would claim a syntax error in any
1430 header after the first one which had an odd number of characters
1433 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1434 callout verification, Exim cached a "reject" for the entire domain. This
1435 is correct for most verifications, but it is not correct for a recipient
1436 verification with use_sender or use_postmaster set, because in that case
1437 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1438 case of MAIL FROM:<> rejection from other early rejections (e.g.
1439 rejection of HELO). When verifying a recipient using a non-null MAIL
1440 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1441 Whatever the result of the callout, the value of the domain cache is
1442 left unchanged (for any other kind of callout, getting as far as trying
1443 RCPT means that the domain itself is ok).
1445 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
1448 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
1449 manifest itself as EPIPE rather than ECONNECT. When tidying away a
1450 session, the daemon ignores ECONNECT errors and logs others; it now
1451 ignores EPIPE as well.
1453 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
1454 (quoted-printable decoding).
1456 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
1457 later the small subsequent patch to fix an introduced bug.
1459 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
1461 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
1463 PH/08 An error is now given if message_size_limit is specified negative.
1465 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
1466 to be given (somewhat) arbitrary names.
1468 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
1471 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
1472 miscellaneous code fixes
1474 PH/10 Added the log_reject_target ACL modifier to specify where to log
1477 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
1478 hostname. This is wrong, because it relates to the incoming message (and
1479 probably the interface on which it is arriving) and not to the outgoing
1480 callout (which could be using a different interface). This has been
1481 changed to use the value of the helo_data option from the smtp transport
1482 instead - this is what is used when a message is actually being sent. If
1483 there is no remote transport (possible with a router that sets up host
1484 addresses), $smtp_active_hostname is used.
1486 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
1487 tweaks were necessary in order to get it to work (see also 21 below):
1488 (a) The code assumed that strncpy() returns a negative number on buffer
1489 overflow, which isn't the case. Replaced with Exim's string_format()
1491 (b) There were several signed/unsigned issues. I just did the minimum
1492 hacking in of casts. There is scope for a larger refactoring.
1493 (c) The code used strcasecmp() which is not a standard C function.
1494 Replaced with Exim's strcmpic() function.
1495 (d) The code set only $1; it now sets $auth1 as well.
1496 (e) A simple test gave the error "authentication client didn't specify
1497 service in request". It would seem that Dovecot has changed its
1498 interface. Fortunately there's a specification; I followed it and
1499 changed what the client sends and it appears to be working now.
1501 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
1504 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
1505 address A is aliased to B and C, where B exists and C does not. Without
1506 -v the output is "A verified" because verification stops after a
1507 successful redirection if more than one address is generated. However,
1508 with -v the child addresses are also verified. Exim was outputting "A
1509 failed to verify" and then showing the successful verification for C,
1510 with its parentage. It now outputs "B failed to verify", showing B's
1511 parentage before showing the successful verification of C.
1513 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
1514 look up a TXT record in a specific list after matching in a combined
1517 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
1518 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
1519 they consult the DNS. I had assumed they would set it the way they
1520 wanted; and indeed my experiments on Linux seem to show that in some
1521 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
1522 To be on the safe side, however, I have now made the interface to
1523 host_find_byname() similar to host_find_bydns(), with an argument
1524 containing the DNS resolver options. The host_find_byname() function now
1525 sets these options at its start, just as host_find_bydns() does. The smtp
1526 transport options dns_qualify_single and dns_search_parents are passed to
1527 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
1528 of host_find_byname() use the default settings of RES_DEFNAMES
1529 (qualify_single) but not RES_DNSRCH (search_parents).
1531 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
1532 spool_read_header() do less string testing, by means of a preliminary
1533 switch on the second character of optional "-foo" lines. (This is
1534 overdue, caused by the large number of possibilities that now exist.
1535 Originally there were few.) While I was there, I also converted the
1536 str(n)cmp tests so they don't re-test the leading "-" and the first
1537 character, in the hope this might squeeze out yet more improvement.
1539 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
1540 flag allowing group syntax was set by the header_syntax check but not
1541 turned off, possible causing trouble later; (2) The flag was not being
1542 set at all for the header_verify test, causing "group"-style headers to
1543 be rejected. I have now set it in this case, and also caused header_
1544 verify to ignore an empty address taken from a group. While doing this, I
1545 came across some other cases where the code for allowing group syntax
1546 while scanning a header line wasn't quite right (mostly, not resetting
1547 the flag correctly in the right place). These bugs could have caused
1548 trouble for malformed header lines. I hope it is now all correct.
1550 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
1551 with the "reply" argument non-NULL. The code, however (which originally
1552 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
1553 but it didn't always do it. This confused somebody who was copying the
1554 code for some other use. I have removed all the tests.
1556 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
1557 feature that was used to support insecure browsers during the U.S. crypto
1558 embargo. It requires special client support, and Exim is probably the
1559 only MTA that supported it -- and would never use it because real RSA is
1560 always available. This code has been removed, because it had the bad
1561 effect of slowing Exim down by computing (never used) parameters for the
1562 RSA_EXPORT functionality.
1564 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
1565 authenticator to fail if there's a tab character in the incoming data
1566 (there should never be unless someone is messing about, as it's supposed
1567 to be base64-encoded). Also added, on Timo's advice, the "secured" option
1568 if the connection is using TLS or if the remote IP is the same as the
1569 local IP, and the "valid-client-cert option" if a client certificate has
1572 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
1573 authenticators. This can be used for authorization after authentication
1574 succeeds. (In the case of plaintext, it servers for both authentication
1577 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
1578 if any retry times were supplied.
1580 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
1581 connection, where there is no client IP address to check. In this
1582 situation, the verify now always succeeds.
1584 PH/25 Applied John Jetmore's -Mset patch.
1586 PH/26 Added -bem to be like -Mset, but loading a message from a file.
1588 PH/27 In a string expansion for a processed (not raw) header when multiple
1589 headers of the same name were present, leading whitespace was being
1590 removed from all of them, but trailing whitespace was being removed only
1591 from the last one. Now trailing whitespace is removed from each header
1592 before concatenation. Completely empty headers in a concatenation (as
1593 before) are ignored.
1595 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
1596 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
1598 PH/29 [Removed. This was a change that I later backed out, and forgot to
1599 correct the ChangeLog entry (that I had efficiently created) before
1600 committing the later change.]
1602 PH/30 Exim was sometimes attempting to deliver messages that had suffered
1603 address errors (4xx response to RCPT) over the same connection as other
1604 messages routed to the same hosts. Such deliveries are always "forced",
1605 so retry times are not inspected. This resulted in far too many retries
1606 for the affected addresses. The effect occurred only when there were more
1607 hosts than the hosts_max_try setting in the smtp transport when it had
1608 the 4xx errors. Those hosts that it had tried were not added to the list
1609 of hosts for which the message was waiting, so if all were tried, there
1610 was no problem. Two fixes have been applied:
1612 (i) If there are any address or message errors in an SMTP delivery, none
1613 of the hosts (tried or untried) are now added to the list of hosts
1614 for which the message is waiting, so the message should not be a
1615 candidate for sending over the same connection that was used for a
1616 successful delivery of some other message. This seems entirely
1617 reasonable: after all the message is NOT "waiting for some host".
1618 This is so "obvious" that I'm not sure why it wasn't done
1619 previously. Hope I haven't missed anything, but it can't do any
1620 harm, as the worst effect is to miss an optimization.
1622 (ii) If, despite (i), such a delivery is accidentally attempted, the
1623 routing retry time is respected, so at least it doesn't keep
1624 hammering the server.
1626 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
1627 in ${readsocket because some servers need this prod.
1629 PH/32 Added some extra debug output when updating a wait-xxx database.
1631 PH/33 The hint "could be header name not terminated by colon", which has been
1632 given for certain expansion errors for a long time, was not being given
1633 for the ${if def:h_colon_omitted{... case.
1635 PH/34 The spec says: "With one important exception, whenever a domain list is
1636 being scanned, $domain contains the subject domain." There was at least
1637 one case where this was not true.
1639 PH/35 The error "getsockname() failed: connection reset by peer" was being
1640 written to the panic log as well as the main log, but it isn't really
1641 panic-worthy as it just means the connection died rather early on. I have
1642 removed the panic log writing for the ECONNRESET error when getsockname()
1645 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
1646 runs only) independently of the message's sender address. This meant
1647 that, if the 4xx error was in fact related to the sender, a different
1648 message to the same recipient with a different sender could confuse
1649 things. In particualar, this can happen when sending to a greylisting
1650 server, but other circumstances could also provoke similar problems.
1651 I have changed the default so that the retry time for these errors is now
1652 based a combination of the sender and recipient addresses. This change
1653 can be overridden by setting address_retry_include_sender=false in the
1656 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
1657 remote server are returned as part of bounce messages. This was not
1658 happening for LMTP over a pipe (the lmtp transport), but now it is the
1659 same for both kinds of LMTP.
1661 PH/38 Despite being documented as not happening, Exim was rewriting addresses
1662 in header lines that were in fact CNAMEs. This is no longer the case.
1664 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
1665 and queue runs started by the daemon processed all messages. This has
1666 been fixed so that -R and -S can now usefully be given with -q<time>.
1668 PH/40 Import PCRE release 6.7 (fixes some bugs).
1670 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
1672 PH/42 Give an error if -q is specified more than once.
1674 PH/43 Renamed the variables $interface_address and $interface_port as
1675 $received_ip_address and $received_port, to make it clear that these
1676 values apply to message reception, and not to the outgoing interface when
1677 a message is delivered. (The old names remain recognized, of course.)
1679 PH/44 There was no timeout on the connect() call when using a Unix domain
1680 socket in the ${readsocket expansion. There now is.
1682 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
1683 be meaningful with "accept".
1685 SC/01 Eximstats V1.43
1686 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
1688 SC/02 Eximstats V1.44
1689 Use a glob alias rather than an array ref in the generated
1690 parser. This improves both readability and performance.
1692 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
1693 Collect SpamAssassin and rejection statistics.
1694 Don't display local sender or destination tables unless
1695 there is data to show.
1696 Added average volumes into the top table text output.
1698 SC/04 Eximstats V1.46
1699 Collect data on the number of addresses (recipients)
1700 as well as the number of messages.
1702 SC/05 Eximstats V1.47
1703 Added 'Message too big' to the list of mail rejection
1704 reasons (thanks to Marco Gaiarin).
1706 SC/06 Eximstats V1.48
1707 Mainlog lines which have GMT offsets and are too short to
1708 have a flag are now skipped.
1710 SC/07 Eximstats V1.49 (Alain Williams)
1711 Added the -emptyok flag.
1713 SC/08 Eximstats V1.50
1714 Fixes for obtaining the IP address from reject messages.
1716 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
1717 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
1718 whitesspace changes from 4.64-PH/27
1720 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
1723 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
1724 are found, allow negative numbers in numeric criteria)
1726 JJ/06 exipick.20061117.2, added new $message_body_missing variable
1728 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
1729 to match changes made in 4.64-PH/43
1731 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
1733 PH/47 Put in an explicit test for a DNS lookup of an address record where the
1734 "domain" is actually an IP address, and force a failure. This locks out
1735 those revolvers/nameservers that support "A-for-A" lookups, in
1736 contravention of the specifications.
1738 PH/48 When a host name was looked up from an IP address, and the subsequent
1739 forward lookup of the name timed out, the host name was left in
1740 $sender_host_name, contrary to the specification.
1742 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
1743 restricted to single-key lookups, Exim was not diagnosing an error if
1744 * or *@ was used with a query-style lookup.
1746 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
1748 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
1749 long ago, but noone interested enough thought of it. Let's just say that
1750 the "1.1" means that there are some new functions that weren't there at
1751 some point in the past.
1753 PH/51 Error processing for expansion failure of helo_data from an smtp
1754 transport during callout processing was broken.
1756 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
1757 tested/used via the -bh/-bhc/-bs options.
1759 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
1760 bug, fixed in subsequent PCRE releases).
1762 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
1763 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
1765 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
1771 SC/01 Use a glob alias rather than an array ref in eximstats generated
1772 parser. This improves both readability and performance.
1774 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
1775 Don't display local sender or destination tables in eximstats unless
1776 there is data to show.
1777 Added average volumes into the eximstats top table text output.
1779 SC/03 Collect data on the number of addresses (recipients) as well
1780 as the number of messages in eximstats.
1782 TF/01 Correct an error in the documentation for the redirect router. Exim
1783 does (usually) call initgroups() when daemonizing.
1785 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
1786 with consistent privilege compared to when running as a daemon.
1788 TF/03 Note in the spec that $authenticated_id is not set for local
1789 submissions from trusted users.
1791 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
1792 Thanks to Dean Brooks <dean@iglou.com> for the patch.
1794 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
1795 by adding some example configuration directives to the default
1796 configuration file. A little bit of work is required to uncomment the
1797 directives and define how usernames and passwords are checked, but
1798 there is now a framework to start from.
1800 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
1801 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
1802 without this. I don't know how relevant this is to other LDAP libraries.
1804 PH/02 Add the verb name to the "unknown ACL verb" error.
1806 PH/03 Magnus Holmgren's patch for filter_prepend_home.
1808 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
1810 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
1811 directory not expanded when it should be if an expanded home directory
1812 was set for the address (which is overridden by the transport).
1814 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
1817 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
1818 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
1819 because it is too late at that time, and has no effect.
1821 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
1822 security issue with \' (bugzilla #107). I could not use the
1823 PQescapeStringConn() function, because it needs a PGconn value as one of
1826 PH/08 When testing addresses using -bt, indicate those final addresses that
1827 are duplicates that would not cause an additional delivery. At least one
1828 person was confused, thinking that -bt output corresponded to deliveries.
1829 (Suppressing duplicates isn't a good idea as you lose the information
1830 about possibly different redirections that led to the duplicates.)
1832 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
1833 systems where poll() doesn't work, in particular OS X.
1835 PH/10 Added more information to debugging output for retry time not reached.
1837 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
1838 operations in malware.c.
1840 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
1843 PH/13 If write_rejectlog was set false when logging was sent to syslog with
1844 syslog_duplication set false, log lines that would normally be written
1845 both the the main log and to the reject log were not written to syslog at
1848 PH/14 In the default configuration, change the use of "message" in ACL warn
1849 statements to "add_header".
1851 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
1852 not followed by a command (e.g. "seen endif").
1854 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
1855 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
1858 PH/17 Added extra conditions to the default value of delay_warning_condition
1862 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
1863 { match{$h_precedence:}{(?i)bulk|list|junk} } \
1864 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
1867 The Auto-Submitted: and various List- headers are standardised, whereas I
1868 don't think Precedence: ever was.
1870 PH/18 Refactored debugging code in route_finduser() to show more information,
1871 in particular, the error code if getpwnam() issues one.
1873 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
1874 This is apparently needed in addition to the PH/07 change above to avoid
1875 any possible encoding problems.
1877 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
1878 but not after initializing Perl.
1880 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
1881 output them only if debugging. By default they are written stderr,
1882 apparently, which is not desirable.
1884 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
1887 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
1890 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1892 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1893 authenticated or an ident call has been made. Suppress the default
1894 values for $authenticated_id and $authenticated_sender (but permit -oMai
1895 and -oMas) when testing with -bh.
1897 PH/24 Re-jigged the order of the tests in the default configuration so that the
1898 tests for valid domains and recipients precede the DNS black list and CSA
1899 tests, on the grounds that those ones are more expensive.
1901 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1902 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1903 This bug exists in every version of Exim that I still have, right back to
1906 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1907 However, an attempt to turn on case-sensitivity in a regex key by
1908 including (?-i) didn't work because the subject string was already
1909 lowercased, and the effects were non-intuitive. It turns out that a
1910 one-line patch can be used to allow (?-i) to work as expected.
1916 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1917 other effects) broke the use of negated acl sub-conditions.
1919 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1922 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1923 "Deny" causes Exim to reject the incoming connection with a 554 error.
1924 Unfortunately, if there is a major crisis, such as a disk failure,
1925 tcp-wrappers gives "deny", whereas what one would like would be some
1926 kind of temporary error. A kludge has been added to help with this.
1927 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1928 554 error is used if errno is still zero or contains ENOENT (which occurs
1929 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1932 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1934 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1935 errors. Otherwise a message that provokes a temporary error (when other
1936 messages do not) can cause a whole host to time out.
1938 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1939 addresses were routed directly to files or pipes from a redirect router.
1940 File deliveries just didn't batch; pipe deliveries might have suffered
1943 PH/06 A failure to get a lock for a hints database would erroneously always say
1944 "Failed to get write lock", even when it was really a read lock.
1946 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1947 of 0600. This has been changed to use the value of the lockfile_mode
1948 option (which defaults to 0600).
1950 PH/08 Applied small patch from the Sieve maintainer.
1952 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1953 folder from quota calculations, a direct delivery into this folder messed
1954 up the contents of the maildirsize file. This was because the regex was
1955 used only to exclude .Trash (or whatever) when the size of the mailbox
1956 was calculated. There was no check that a delivery was happening into an
1957 excluded directory. This bug has been fixed by ignoring all quota
1958 processing for deliveries into excluded directories.
1960 PH/10 Added the maildirfolder_create_regex option to appendfile.
1966 PH/01 The code for finding all the local interface addresses on a FreeBSD
1967 system running IPv6 was broken. This may well have applied to all BSD
1968 systems, as well as to others that have similar system calls. The broken
1969 code found IPv4 interfaces correctly, but gave incorrect values for the
1970 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1971 that it would not match correctly against @[] and not recognize the IPv6
1974 PH/02 The ipliteral router was not recognizing addresses of the form user@
1975 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1977 PH/03 Added disable_ipv6.
1979 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1980 decoded form, because it is most often used to construct To: headers
1981 lines in autoreplies, and the decoded form may well be syntactically
1982 invalid. However, $reply_address has leading white space removed, and all
1983 newlines turned into spaces so that the autoreply transport does not
1986 PH/05 If group was specified without a user on a router, and no group or user
1987 was specified on a transport, the group from the router was ignored.
1989 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1990 for visible compile-time settings that can be used to change these
1991 numbers, for those that want even more. Backwards compatibility with old
1992 spool files has been maintained. However, going back to a previous Exim
1993 release will lost any variables that are in spool files.
1995 PH/07 Two small changes when running in the test harness: increase delay when
1996 passing a TCP/IP connection to a new process, in case the original
1997 process has to generate a bounce, and remove special handling of
1998 127.0.0.2 (sic), which is no longer necessary.
2000 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2001 be the same on different OS.
2003 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2006 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2007 whether --show-vars was specified or not
2009 JJ/02 exipick: Added support for new ACL variable spool format introduced
2012 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2013 syntactically invalid From: or Reply-to: line, and a filter used this to
2014 generate an autoreply, and therefore failed to obtain an address for the
2015 autoreply, Exim could try to deliver to a non-existent relative file
2016 name, causing unrelated and misleading errors. What now happens is that
2017 it logs this as a hard delivery error, but does not attempt to create a
2020 PH/11 The exinext utility has a -C option for testing purposes, but although
2021 the given file was scanned by exinext itself; it wasn't being passed on
2022 when Exim was called.
2024 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2025 an end-of-file indication when reading a command response.
2027 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2028 compiled. In many other places in Exim, IPv6 addresses are always
2029 recognized, so I have changed this. It also means that IPv4 domain
2030 literals of the form [IPV4:n.n.n.n] are now always recognized.
2032 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2033 used if the router is not running as root, for example, when verifying at
2034 ACL time, or when using -bh. The debugging output from this situation was
2035 non-existent - all you got was a failure to exec. I have made two
2038 (a) Failures to set uid/gid, the current directory, or a process leader
2039 in a subprocess such as that created by queryprogram now generate
2040 suitable debugging ouput when -d is set.
2042 (b) The queryprogram router detects when it is not running as root,
2043 outputs suitable debugging information if -d is set, and then runs
2044 the subprocess without attempting to change uid/gid.
2046 PH/15 Minor change to Makefile for building test_host (undocumented testing
2049 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2050 additional section of a DNS packet that returns MX or SRV records.
2051 Instead, it always explicitly searches for A/AAAA records. This avoids
2052 major problems that occur when a DNS server includes only records of one
2053 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2054 fixed another bug: if SRV records were looked up and the corresponding
2055 address records were *not* found in the additional section, the port
2056 values from the SRV records were lost.
2058 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2059 using the correct key (the original address) when searching the retry
2060 rules in order to find which one to use for generating the retry hint.
2062 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2063 adding the default one. Similarly, if it contains a Reply-To: header, the
2064 errors_reply_to option, if set, is not used.
2066 PH/19 When calculating a retry time, Exim used to measure the "time since
2067 failure" by looking at the "first failed" field in the retry record. Now
2068 it does not use this if it is later than than the arrival time of the
2069 message. Instead it uses the arrival time. This makes for better
2070 behaviour in cases where some deliveries succeed, thus re-setting the
2071 "first failed" field. An example is a quota failure for a huge message
2072 when small messages continue to be delivered. Without this change, the
2073 "time since failure" will always be short, possible causing more frequent
2074 delivery attempts for the huge message than are intended.
2075 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2077 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2078 $1, $2, $3) because the numerical variables can be reset during some
2079 expansion items (e.g. "match"), thereby losing the authentication data.
2081 PH/21 Make -bV show the size of off_t variables so that the test suite can
2082 decide whether to run tests for quotas > 2G.
2084 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2085 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2086 filecount value is greater than 2G or if a quota value is greater than 2G
2087 on a system where the size of off_t is not greater than 4, a panic error
2090 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2091 never match. The debug and -bh output now contains an explicit error
2092 message indicating a malformed IPv4 address or mask.
2094 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2095 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2096 PH/23 above applies.
2098 PH/25 Do not write to syslog when running in the test harness. The only
2099 occasion when this arises is a failure to open the main or panic logs
2100 (for which there is an explicit test).
2102 PH/26 Added the /no_tell option to "control=freeze".
2104 PH/27 If a host name lookup failed very early in a connection, for example, if
2105 the IP address matched host_lookup and the reverse lookup yielded a name
2106 that did not have a forward lookup, an error message of the form "no IP
2107 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2108 could be logged. Now it outputs the IP address instead of "NULL".
2110 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2111 allows the sender and the authenticated sender to be set when
2112 submitting a message from within Exim. Since child_open_exim() is
2113 documented for local_scan(), the new function should be too.
2115 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2116 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2117 results in an empty string is now treated as unset.
2119 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2121 PH/31 Added sender_verify_fail logging option.
2123 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2124 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2125 tidied the source and removed it altogether.
2127 PH/33 When a queue run was abandoned because the load average was too high, a
2128 log line was always written; now it is written only if the queue_run log
2129 selector is set. In addition, the log line for abandonment now contains
2130 information about the queue run such as the pid. This is always present
2131 in "start" and "stop" lines but was omitted from the "abandon" line.
2133 PH/34 Omit spaces between a header name and the colon in the error message that
2134 is given when verify = headers_syntax fails (if there are lots of them,
2135 the message gets confusing).
2137 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2138 names, as there are now some PTR records that contain slashes. This check
2139 is only to protect against broken name servers that fall over on strange
2140 characters, so the fact that it applies to all lookups doesn't matter.
2142 PH/36 Now that the new test suite is complete, we can remove some of the
2143 special code in Exim that was needed for the old test suite. For example,
2144 sorting DNS records because real resolvers return them in an arbitrary
2145 order. The new test suite's fake resolver always returns records in the
2148 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2149 bounces) except when queue_only is set, to avoid logging races between
2150 the different processes.
2152 PH/38 Panic-die if .include specifies a non-absolute path.
2154 PH/39 A tweak to the "H" retry rule from its user.
2156 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2157 a label. They prevented compilation on older perls.
2159 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2160 a warning to be raised on newish perls.
2162 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2163 on queue. Changes to match documented behaviour of showing count of
2164 messages matching specified criteria.
2166 PH/40 Changed the default ident timeout from 30s to 5s.
2168 PH/41 Added support for the use of login_cap features, on those BSD systems
2169 that have them, for controlling the resources used by pipe deliveries.
2171 PH/42 The content-scanning code uses fopen() to create files in which to put
2172 message data. Previously it was not paying any attention to the mode of
2173 the files. Exim runs with umask(0) because the rest of the code creates
2174 files with open(), and sets the required mode explicitly. Thus, these
2175 files were ending up world-writeable. This was not a big issue, because,
2176 being within the spool directory, they were not world-accessible. I have
2177 created a function called modefopen, which takes an additional mode
2178 argument. It sets umask(777), creates the file, chmods it to the required
2179 mode, then resets the umask. All the relevant calls to fopen() in the
2180 content scanning code have been changed to use this function.
2182 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2183 to 24 hours. This avoids potential overflow problems when processing G
2184 and H retry rules. I suspect nobody ever tinkers with this value.
2186 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2188 PH/45 When the plaintext authenticator is running as a client, the server's
2189 challenges are checked to ensure they are valid base64 strings. By
2190 default, the authentication attempt is cancelled if an invalid string is
2191 received. Setting client_ignore_invalid_base64 true ignores these errors.
2192 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2193 they are received. Thus, the responses can be made to depend on the
2194 challenges. If an invalid string is ignored, an empty string is placed in
2197 PH/46 Messages that are created by the autoreply transport now contains a
2198 References: header, in accordance with RFCs 2822 and 3834.
2200 PH/47 Added authenticated_sender_force to the smtp transport.
2202 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2204 PH/49 Installed latest patch from the Sieve maintainer.
2206 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2207 was also set, the appendfile transport was unnecessarily scanning a
2208 directory of message files (e.g. for maildir delivery) to find the count
2209 of files (along with the size), even though it did not need this
2210 information. It now does the scan only if it needs to find either the
2211 size of the count of files.
2213 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2215 PH/52 Two bugs concerned with error handling when the smtp transport is
2218 (i) Exim was not creating retry information for temporary errors given
2219 for individual recipients after the DATA command when the smtp transport
2220 was used in LMTP mode. This meant that they could be retried too
2221 frequently, and not timed out correctly.
2223 (ii) Exim was setting the flag that allows error details to be returned
2224 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2225 recipients that were returned after the DATA command.
2227 PH/53 This is related to PH/52, but is more general: for any failing address,
2228 when detailed error information was permitted to be returned to the
2229 sender, but the error was temporary, then after the final timeout, only
2230 "retry timeout exceeded" was returned. Now it returns the full error as
2231 well as "retry timeout exceeded".
2233 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2234 do this, and (what is worse) MTAs that accept it.
2236 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2237 will now be deprecated.
2239 PH/56 New os.c-cygwin from the Cygwin maintainer.
2241 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2242 formats (previously only available in exim formats via -bpr, -bpru,
2243 and -bpra. Now also available in native and exiqgrep formats)
2245 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2246 with very large, slow to parse queues
2248 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2250 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2252 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2253 responses to authentication challenges, though it was showing the
2254 challenges; (ii) I've removed the CR characters from the debug output for
2257 PH/58 Allow for the insertion of a newline as well as a space when a string
2258 is turned into more than one encoded-word during RFC 2047 encoding. The
2259 Sieve code now uses this.
2261 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2262 data_4xx, lost_connection, tls_required.
2264 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2265 message was being sent as an SMTP response.
2267 PH/61 Add -l and -k options to exicyclog.
2269 PH/62 When verifying, if an address was redirected to one new address, so that
2270 verification continued, and the new address failed or deferred after
2271 having set something in $address_data, the value of $address_data was not
2272 passed back to the ACL. This was different to the case when no
2273 redirection occurred. The value is now passed back in both cases.
2275 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2276 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2277 use login_cap.h, so on its own it isn't the distinguishing feature. The
2278 new name refers directly to the setclassresources() function.
2280 PH/65 Added configuration files for NetBSD3.
2282 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2284 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2285 is preferred over IPv4.
2287 PH/68 The bounce_return_message and bounce_return_body options were not being
2288 honoured for bounces generated during the reception of non-SMTP messages.
2289 In particular, this applied to messages rejected by the ACL. This bug has
2290 been fixed. However, if bounce_return_message is true and bounce_return_
2291 body is false, the headers that are returned for a non-SMTP message
2292 include only those that have been read before the error was detected.
2293 (In the case of an ACL rejection, they have all been read.)
2295 PH/69 The HTML version of the specification is now built in a directory called
2296 spec_html instead of spec.html, because the latter looks like a path with
2297 a MIME-type, and this confuses some software.
2299 PH/70 Catch two compiler warnings in sieve.c.
2301 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2302 function verify_get_ident() calls ip_connect() to connect a socket, but
2303 if the "connect()" function timed out, ip_connect() used to close the
2304 socket. However, verify_get_ident() also closes the socket later, and in
2305 between Exim writes to the log, which may get opened at this point. When
2306 the socket was closed in ip_connect(), the log could get the same file
2307 descriptor number as the socket. This naturally causes chaos. The fix is
2308 not to close the socket in ip_connect(); the socket should be closed by
2309 the function that creates it. There was only one place in the code where
2310 this was missing, in the iplookup router, which I don't think anybody now
2311 uses, but I've fixed it anyway.
2313 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2314 well as to direct DNS lookups. Otherwise the handling of names in host
2315 lists is inconsistent and therefore confusing.
2321 PH/01 Two changes to the default runtime configuration:
2323 (1) Move the checks for relay_from_hosts and authenticated clients from
2324 after to before the (commented out) DNS black list checks.
2326 (2) Add control=submission to the relay_from_hosts and authenticated
2327 clients checks, on the grounds that messages accepted by these
2328 statements are most likely to be submissions.
2330 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2332 (1) Generate an error if the third argument for the ${prvs expansion is
2335 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2338 (3) Reset the variables that are obtained from the first argument of
2339 ${prvscheck and used in the second argument before leaving the code,
2340 because their memory is reclaimed, so using them afterwards may do
2343 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2344 one (it's much easier than Tom thought :-).
2346 (5) Because of (4), we can now allow for the use of $prvscheck_result
2347 inside the third argument.
2349 PH/03 For some reason, the default setting of PATH when running a command from
2350 a pipe transport was just "/usr/bin". I have changed it to
2353 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2354 anything to be listed in the output from -bV.
2356 PH/05 When a filter generated an autoreply, the entire To: header line was
2357 quoted in the delivery log line, like this:
2359 => >A.N.Other <ano@some.domain> <original@ddress> ...
2361 This has been changed so that it extracts the operative address. There
2362 may be more than one such address. If so, they are comma-separated, like
2365 => >ano@some.domain,ona@other.domain <original@ddress> ...
2367 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2368 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2369 not being looked up in the rDNS to get a host name, Exim was showing the
2370 IP address twice in Received: lines, even though the IP addresses were
2371 identical. For example:
2373 Received: from [1.2.3.4] (helo=[1.2.3.4])
2375 However, if the real host name was known, it was omitting the HELO data
2376 if it matched the actual IP address. This has been tidied up so that it
2377 doesn't show the same IP address twice.
2379 PH/07 When both +timestamp and +memory debugging was on, the value given by
2380 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2381 called by the debug printing, thereby overwriting the timestamp buffer.
2382 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2384 PH/08 When the original message was included in an autoreply transport, it
2385 always said "this is a copy of the message, including all the headers",
2386 even if body_only or headers_only was set. It now gives an appropriate
2389 PH/09 Applied a patch from the Sieve maintainer which:
2391 o fixes some comments
2392 o adds the (disabled) notify extension core
2393 o adds some debug output for the result of if/elsif tests
2394 o points to the current vacation draft in the documentation
2395 and documents the missing references header update
2399 o fixes a bug in processing the envelope test (when testing
2400 multiple envelope elements, the last element determinted the
2403 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2404 Electronic Mail") by including:
2406 Auto-submitted: auto-generated
2408 in the messages that it generates (bounce messages and others, such as
2409 warnings). In the case of bounce messages for non-SMTP mesages, there was
2410 also a typo: it was using "Auto_submitted" (underscore instead of
2411 hyphen). Since every message generated by Exim is necessarily in response
2412 to another message, thes have all been changed to:
2414 Auto-Submitted: auto-replied
2416 in accordance with these statements in the RFC:
2418 The auto-replied keyword:
2420 - SHOULD be used on messages sent in direct response to another
2421 message by an automatic process,
2423 - MUST NOT be used on manually-generated messages,
2425 - MAY be used on Delivery Status Notifications (DSNs) and Message
2426 Disposition Notifications (MDNs),
2428 - MUST NOT be used on messages generated by automatic or periodic
2429 processes, except for messages which are automatic responses to
2432 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2433 to the default Received: header definition.
2435 PH/12 Added log selector acl_warn_skipped (default on).
2437 PH/13 After a successful wildlsearch lookup, discard the values of numeric
2438 variables because (a) they are in the wrong storage pool and (b) even if
2439 they were copied, it wouldn't work properly because of the caching.
2441 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2442 checking when decoding. Apparently there are clients that generate
2443 overlong encoded strings. Why am I not surprised?
2445 PH/15 If the first argument of "${if match_address" was not empty, but did not
2446 contain an "@" character, Exim crashed. Now it writes a panic log message
2447 and treats the condition as false.
2449 PH/16 In autoreply, treat an empty string for "once" the same as unset.
2451 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
2452 extension "envelope-auth". The code is finished and in agreement with
2453 other implementations, but there is no documentation so far and in fact,
2454 nobody wrote the draft yet. This extension is currently #undef'ed, thus
2455 not changing the active code.
2457 Print executed "if" and "elsif" statements when debugging is used. This
2458 helps a great deal to understand what a filter does.
2460 Document more things not specified clearly in RFC3028. I had all this
2461 sorted out, when out of a sudden new issues came to my mind. Oops."
2463 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
2466 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
2467 canonical form (as documented). However, after a host name lookup from
2468 the IP address, check_host() was doing a simple string comparison with
2469 addresses acquired from the DNS when checking that the found name did
2470 have the original IP as one of its addresses. Since any found IPv6
2471 addresses are likely to be in abbreviated form, the comparison could
2472 fail. Luckily, there already exists a function for doing the comparison
2473 by converting both addresses to binary, so now that is used instead of
2474 the text comparison.
2476 PH/20 There was another similar case to PH/19, when a complete host name was
2477 given in a host list; looking up its IP address could give an abbreviated
2478 form, whereas the current host's name might or might not be abbreviated.
2479 The same fix has been applied.
2485 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
2486 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
2489 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
2490 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
2492 PH/03 Typo: missing ".o" in src/pcre/Makefile.
2494 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
2495 header line, restrict the check to what is listed in RFCs 2369 and 2929.
2496 Also, for "Auto-Submitted", treat anything other than "no" as
2497 non-personal, in accordance with RFC 3834. (Previously it treated
2498 anything starting "auto-" as non-personal.)
2500 TF/01 The control=submission/name=... option had a problem with syntax
2501 errors if the name included a slash character. The /name= option
2502 now slurps the rest of the string, so it can include any characters
2503 but it must come last in the list of options (after /sender_retain
2506 PH/05 Some modifications to the interface to the fake nameserver for the new
2514 TK/01 Added the "success_on_redirect" address verification option. See
2515 NewStuff for rationale and an example.
2517 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
2519 PH/02 Patch to exigrep to allow it to work on syslog lines.
2521 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
2522 fread() to skip over the body file's header line, because in Cygwin the
2523 header line is locked and is inaccessible.
2525 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
2526 co-exist for some time) to make it clear that it is the Exim ID that is
2527 referenced, not the Message-ID: header line.
2529 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
2530 string_format() function, because snprintf() does not exist on all
2533 PH/06 The use of forbid_filter_existstest now also locks out the use of the
2534 ${stat: expansion item.
2536 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
2537 protocol synchronization error", to keep the pedants happy.
2539 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
2540 well as for IRIX systems, when gcc is being used. See the host.c source
2543 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
2545 PH/10 Named domain lists were not working if used in a queue_smtp_domains
2548 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
2549 transport and to the smtp transport in LMTP mode.
2551 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
2553 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
2554 run a filter in a subprocess. This could lead to confusion in subsequent
2555 lookups in the parent process. There should also be a search_tidyup() at
2556 the end of the subprocess.
2558 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
2559 only if the host matched helo_try_verify_hosts, which caused the
2560 verification to occur when the EHLO/HELO command was issued. The ACL just
2561 tested the remembered result. Now, if a previous verification attempt has
2562 not happened, "verify = helo" does it there and then.
2564 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
2566 TK/03 Fix log output including CR from clamd.
2568 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
2569 exist provoked a memory error which could cause a segfault.
2571 PH/15 Installed PCRE 6.2
2573 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
2575 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
2576 of the problem. Specifically, suggested +O2 rather than +O1 for the
2579 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
2581 PH/20 If a delivery was routed to a non-standard port by means of an SRV
2582 record, the port was not correctly logged when the outgoing_port log
2583 selector was set (it logged the transort's default port).
2585 PH/21 Added support for host-specific ports to manualroute, queryprogram,
2586 fallback_hosts, and "hosts" in the smtp transport.
2588 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
2589 host errors such as "Connection refused".
2591 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
2592 authentication with radiusclient 0.4.9:
2594 - Error returned from rc_read_config was caught wrongly
2595 - Username/password not passed on to radius server due to wrong length.
2597 The presumption is that some radiusclient API changes for 4.51/PH/17
2598 were not taken care of correctly. The code is still untested by me (my
2599 Linux distribution still has 0.3.2 of radiusclient), but it was
2600 contributed by a Radius user.
2602 PH/24 When doing a callout, the value of $domain wasn't set correctly when
2603 expanding the "port" option of the smtp transport.
2605 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
2606 while reading a MIME header. Thanks to Tom Hughes for a patch.
2608 PH/24 Include config.h inside local_scan.h so that configuration settings are
2611 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
2612 that in an ACL for RCPT (for example), you can examine exactly what was
2615 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
2616 commands, but it was not correctly comparing the address with the actual
2617 client host address. Thus, it would show the EHLO address in Received:
2618 header lines when this was not necessary.
2620 PH/27 Added the % operator to ${eval:}.
2622 PH/28 Exim tries to create and chdir to its spool directory when it starts;
2623 it should be ignoring failures (because with -C, for example, it has lost
2624 privilege). It wasn't ignoring creation failures other than "already
2627 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
2630 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
2631 because an input line was too long, either on its own, or by virtue of
2632 too many continuations, the temporary file was not being removed, and the
2633 return code was incorrect.
2635 PH/31 Missing "BOOL" in function definition in filtertest.c.
2637 PH/32 Applied Sieve patches from the maintainer.
2639 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
2641 PH/33 Added "verify = not_blind".
2643 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
2644 Local/Makefile (with some defaults set). These are used in built scripts
2645 such as exicyclog, but they have never been used in the exim_install
2646 script (though there are many overriding facilities there). I have
2647 arranged that the exim_install script now takes note of these two
2650 PH/35 Installed configuration files for Dragonfly.
2652 PH/36 When a locally submitted message by a trusted user did not contain a
2653 From: header, and the sender address was obtained from -f or from an SMTP
2654 MAIL command, and the trusted user did not use -F to supply a sender
2655 name, $originator_name was incorrectly used when constructing a From:
2656 header. Furthermore, $originator_name was used for submission mode
2657 messages from external hosts without From: headers in a similar way,
2658 which is clearly wrong.
2660 PH/37 Added control=suppress_local_fixups.
2662 PH/38 When log_selector = +received_sender was set, and the addition of the
2663 sender made the log line's construction buffer exactly full, or one byte
2664 less than full, an overflow happened when the terminating "\n" was
2667 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
2668 when the result of a list match is failure because a DNS lookup failed.
2670 PH/40 RM_COMMAND is now used in the building process.
2672 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
2673 the "build-* directories that it finds.
2675 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
2676 address in a domain literal was a prefix of an interface address.
2678 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
2679 when verifying a sender address, unless rewrite_headers is false.
2681 PH/44 Wrote a long comment about why errors_to addresses are verified as
2682 recipients, not senders.
2684 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
2685 the ratelimit ACL was added.
2687 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
2689 PH/46 Added extra information about PostgreSQL errors to the error string.
2691 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
2692 avoiding the need to install special zones in a real server. This is
2693 backwards compatible; if it can't find the fake resolver, it drops back.
2694 Thus, both old and new test suites can be run.
2696 TF/02 Added util/ratelimit.pl
2698 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
2699 clock is set back in time.
2701 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
2702 Candler <B.Candler@pobox.com>.
2704 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
2705 OK for addresses that are the result of redirections.
2707 PH/48 A number of further additions for the benefit of the new test suite,
2708 including a fake gethostbyname() that interfaces to the fake DNS resolver
2711 TF/06 The fix for widen_domains has also been applied to qualify_single and
2712 search_parents which are the other dnslookup options that can cause
2715 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
2718 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
2720 TF/07 Exim produced the error message "an SRV record indicated no SMTP
2721 service" if it encountered an MX record with an empty target hostname.
2722 The message is now "an MX or SRV record indicated no SMTP service".
2724 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
2725 if the DNS of the sending site is misconfigured. This is quite a
2726 common situation. This change restores the behaviour of treating a
2727 helo verification defer as a failure.
2729 PH/51 If self=fail was set on a router, the bounce message did not include the
2730 actual error message.
2736 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
2738 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
2739 command itself ended in error, the underlying message about the transport
2740 filter timeout was being overwritten with the pipe command error. Now the
2741 underlying error message should be appended to the second error message.
2743 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
2745 PH/02 Reduce the amount of output that "make" produces by default. Full output
2746 can still be requested.
2748 PH/03 The warning log line about a condition test deferring for a "warn" verb
2749 was being output only once per connection, rather than after each
2750 occurrence (because it was using the same function as for successful
2751 "warn" verbs). This seems wrong, so I have changed it.
2753 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
2754 it should not have, which might have caused a crash in the right
2755 circumstances, but probably never did.
2757 PH/04 Installed a modified version of Tony Finch's patch to make submission
2758 mode fix the return path as well as the Sender: header line, and to
2759 add a /name= option so that you can make the user's friendly name appear
2762 TF/03 Added the control = fakedefer ACL modifier.
2764 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
2765 Mark Lowes for thorough testing.
2767 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
2769 TK/03 Merged latest SRS patch from Miles Wilton.
2771 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
2772 with the definition in sysexits.h (which is #included earlier).
2773 Fortunately, Exim does not actually use EX_OK. The code used to try to
2774 preserve the sysexits.h value, by assumimg that macro definitions were
2775 scanned for macro replacements. I have been disabused of this notion,
2776 so now the code just undefines EX_OK before #including unistd.h.
2778 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
2779 in the smtp transport. When a block could not be written in a single
2780 write() function, the timeout was being re-applied to each part-write.
2781 This seems wrong - if the receiver was accepting one byte at a time it
2782 would take for ever. The timeout is now adjusted when this happens. It
2783 doesn't have to be particularly precise.
2785 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
2786 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
2788 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
2789 without a domain if the check to <postmaster@domain> fails.
2791 SC/01 Eximstats: added -xls and the ability to specify output files
2792 (patch written by Frank Heydlauf).
2794 SC/02 Eximstats: use FileHandles for outputing results.
2796 SC/03 Eximstats: allow any combination of xls, txt, and html output.
2798 SC/04 Eximstats: fixed display of large numbers with -nvr option
2800 SC/05 Eximstats: fixed merging of reports with empty tables.
2802 SC/06 Eximstats: added the -include_original_destination flag
2804 SC/07 Eximstats: removed tabs and trailing whitespace.
2806 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
2808 TK/06 MBOX spool code: Add real "From " MBOX separator line
2809 so the .eml file is really in mbox format (even though
2810 most programs do not really care). Patch from Alex Miller.
2812 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
2813 The latter is generated from $received_to and is only set if the
2814 message has one envelope recipient. SA can use these headers,
2815 obviously out-of-the-box. Patch from Alex Miller.
2817 PH/08 The ${def test on a variable was returning false if the variable's
2818 value was "0", contrary to what the specification has always said!
2819 The result should be true unless the variable is empty.
2821 PH/09 The syntax error of a character other than { following "${if
2822 def:variable_name" (after optional whitespace) was not being diagnosed.
2823 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
2824 accidental colon was present, for example, could give incorrect results.
2826 PH/10 Tidied the code in a number of places where the st_size field of a stat()
2827 result is used (not including appendfile, where other changes are about
2830 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
2831 This involved changing a lot of size variables from int to off_t. It
2832 should work with maildirs and everything.
2834 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
2835 spamd dying while we are connected to it.
2837 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
2840 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
2841 function for BDB 4.3.
2843 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
2845 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
2846 the sources for PCRE have been re-arranged and more files are now
2849 PH/15 The code I had for printing potentially long long variables in PH/11
2850 above was not the best (it lost precision). The length of off_t variables
2851 is now inspected at build time, and an appropriate printing format (%ld
2852 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
2853 to be "long long int" or "long int". This is needed for the internal
2854 formatting function string_vformat().
2856 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
2857 the configuration file to be ":syslog", then the script "guesses" where
2858 the logs files are, rather than using the compiled in default. In our
2859 case the guess is not the same as the compiled default, so the script
2860 suddenly stopped working when I started to use syslog. The patch checks
2861 to see if log_file_path is "". If so, it attempts to read it from exim
2862 with no configuration file to get the compiled in version, before it
2863 falls back to the previous guessing code."
2865 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
2866 implementing BATV in an Exim configuration. See NewStuff for the gory
2869 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
2870 Makefile that are specific to HP-UX.
2872 PH/18 If the "use_postmaster" option was set for a recipient callout together
2873 with the "random" option, the postmaster address was used as the MAIL
2874 FROM address for the random test, but not for the subsequent recipient
2875 test. It is now used for both.
2877 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
2878 patch removes a few documentation additions to RFC 3028, because the
2879 latest draft now contains them. It adds the new en;ascii-case comparator
2880 and a new error check for 8bit text in MIME parts. Comparator and
2881 require names are now matched exactly. I enabled the subaddress
2882 extension, but it is not well tested yet (read: it works for me)."
2884 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
2885 rework some of the code of TK/09 above to avoid the hardwired use of
2886 "%lld" and "long long". Replaced the call to snprintf() with a call to
2889 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
2890 records point to non-existent hosts", "retry timeout exceeded", and
2891 "retry time not reached for any host after a long failure period".
2893 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2894 experimental DomainKeys support:
2896 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2897 (2) On an error such as an illegally used "control", the wrong name for
2898 the control was given.
2900 These problems did NOT occur unless DomainKeys support was compiled.
2902 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2904 PH/24 Added ${if match_ip condition.
2906 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2907 databases so that it will be absolutely obvious if a crash occurs in the
2908 DB library. This is a regular occurrence (often caused by mis-matched
2911 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2912 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2913 picked up on a user's system that detects such things. There doesn't seem
2914 to be a gcc warning option for this - only an attribute that has to be
2915 put on the function's prototype. It seems that in Fedora Core 4 they have
2916 set this on a number of new functions. No doubt there will be more in due
2919 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2920 specify a transport. However, if an address that was verified by such a
2921 router was the subject of a callout, Exim crashed because it tried to
2922 read the rcpt_include_affixes from the non-existent transport. Now it
2923 just assumes that the setting of that option is false. This bug was
2924 introduced by 4.51/PH/31.
2926 PH/28 Changed -d+all to exclude +memory, because that information is very
2927 rarely of interest, but it makes the output a lot bigger. People tend to
2928 do -d+all out of habit.
2930 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2931 code in os-type was giving problems when libc.so lives in lib64, like on
2934 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2935 aren't the modern standard, and it seems that some systems' include files
2936 don't always have them. Exim was already checking for some of the newer
2937 ones like T_AAAA, and defining it itself. I've added checks for all the
2938 record types that Exim uses.
2940 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2941 not automatically generating a new one, as it is supposed to. This
2942 prevented TLS from working. If the file did exist, but contained invalid
2943 data, a new version was generated, as expected. It was only the case of a
2944 non-existent file that was broken.
2946 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2947 with a change in libdomainkeys > 0.64.
2949 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2950 from DNS. If the selector record carries the flag, it now has
2951 precedence over the domain-wide flag.
2953 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2955 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2956 the use of a port name that isn't defined in /etc/services) occurred, the
2957 message was deferred as in a normal delivery, and thus remained on the
2958 spool, instead of being failed because of the mua_wrapper setting. This
2959 is now fixed, and I tidied up some of the mua_wrapper messages at the
2962 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2963 the messages in a hash of arrays rather than using individual hashes.
2964 This is a bit cleaner and results in dramatic memory savings, albeit
2965 at a slight CPU cost.
2967 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2968 as requested by Marc Sherman.
2970 SC/10 Eximstats: added histograms for user specified patterns as requested
2973 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2975 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2976 fopen() in the content-scanning modules that did not already have it.
2982 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2983 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2985 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2987 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2989 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2990 to test Sieve filters that use "vacation".
2992 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2993 that changes the way the GnuTLS parameters are stored in the cache file.
2994 The new format can be generated externally. For backward compatibility,
2995 if the data in the cache doesn't make sense, Exim assumes it has read an
2996 old-format file, and it generates new data and writes a new file. This
2997 means that you can't go back to an older release without removing the
3000 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3001 work if there are any delivery delays because "one_time" forces the
3002 parent to be marked "delivered", so its unseen clone is never tried
3003 again. For this reason, Exim now forbids the simultaneous setting of
3006 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3007 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3008 entry if you want to know the details. The fix, however, neglected to
3009 consider the case when local delivery batching is involved. The test for
3010 "previously delivered" was not happening when checking to see if an
3011 address could be batched with a previous (undelivered) one; under
3012 certain circumstances this could lead to multiple deliveries to the same
3015 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3016 in its include files, and this causes problems building Exim.
3018 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3019 header_syntax) but Exim was just ignoring anything given after a slash.
3020 In particular, this caused confusion with an attempt to use "verify =
3021 reverse_host_lookup/defer_ok". An error is now given when options are
3022 supplied for verify items that do not have them. (Maybe reverse_host_
3023 lookup should have a defer_ok option, but that's a different point.)
3025 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3026 defined by RFC 821) to 2048, because there were problems with some AUTH
3027 commands, and RFC 1869 says the size should be increased for extended
3028 SMTP commands that take arguments.
3030 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3033 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3034 "unknown" error; now it says that the functionality isn't in the binary.
3036 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3037 an address' error message when a string expansion fails (syntax or
3038 whatever). Otherwise the password may appear in the log. Following change
3039 PH/42 below, there is no longer a chance of it appearing in a bounce
3042 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3044 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3045 of its addresses was ever tried. (Bugzilla bug #2.)
3047 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3048 the result incorrectly in the debug output. (It correctly added a newline
3049 to what was transported.)
3051 TF/01 Added $received_time.
3053 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3054 commented out examples of how to interface to a virus scanner and to
3055 SpamAssassin. Also added commented examples of av_scanner and
3056 spamd_address settings.
3058 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3059 and controls are allowed in which ACLs. There were a couple of minor
3060 errors. Some of the entries in the conditions table (which is a table of
3061 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3062 negation of where the condition IS allowed.
3064 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3066 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3067 header file does not have a version number, so I've had to invent a new
3068 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3069 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3070 radiusclient), but it was contributed by a Radius user.
3072 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3073 files or directories, for OpenSSL.
3075 PH/19 When an Exim process that is running as root has to create an Exim log
3076 file, it does so in a subprocess that runs as exim:exim so as to get the
3077 ownership right at creation (otherwise, other Exim processes might see
3078 the file with the wrong ownership). There was no test for failure of this
3079 fork() call, which would lead to the process getting stuck as it waited
3080 for a non-existent subprocess. Forks do occasionally fail when resources
3081 run out. I reviewed all the other calls to fork(); they all seem to check
3084 PH/20 When checking for unexpected SMTP input at connect time (before writing
3085 the banner), Exim was not dealing correctly with a non-positive return
3086 from the read() function. If the client had disconnected by this time,
3087 the result was a log entry for a synchronization error with an empty
3088 string after "input=" when read() returned zero. If read() returned -1
3089 (an event I could not check), uninitialized data bytes were printed.
3090 There were reports of junk text (parts of files, etc) appearing after
3093 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3095 PH/22 Added support for macro redefinition, and (re)definition in between
3096 driver and ACL definitions.
3098 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3099 forgetting to use the resulting value; it was using the unexpanded value.
3101 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3102 hadn't been configured. The fix is from Juergen Kreileder, who
3103 understands it better than I do:
3105 "Here's what I see happening with three configured cyrus_sasl
3106 authenticators configured (plain, login, cram-md5):
3108 On startup auth_cyrus_sasl_init() gets called for each of these.
3109 This means three calls to sasl_listmech() without a specified mech_list.
3110 => SASL tests which mechs of all available mechs actually work
3111 => three warnings about OTP not working
3112 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3114 With the patch, sasl_listmech() also gets called three times. But now
3115 SASL's mech_list option is set to the server_mech specified in the the
3116 authenticator. Or in other words, the answer from sasl_listmech()
3117 gets limited to just the mech you're testing for (which is different
3119 => the return list contains just 'plain' or 'login', 'cram-md5' or
3120 nothing depending on the value of ob->server_mech.
3122 I've just tested the patch: Authentication still works fine,
3123 unavailable mechs specified in the exim configuration are still
3124 caught, and the auth.log warnings about OTP are gone."
3126 PH/25 When debugging is enabled, the contents of the command line are added
3127 to the debugging output, even when log_selector=+arguments is not
3130 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3131 answer is "GNU", and only if the return is "GNU/something" is the answer
3134 PH/27 $acl_verify_message is now set immediately after the failure of a
3135 verification in an ACL, and so is available in subsequent modifiers. In
3136 particular, the message can be preserved by coding like this:
3138 warn !verify = sender
3139 set acl_m0 = $acl_verify_message
3141 Previously, $acl_verify_message was set only while expanding "message"
3142 and "log_message" when a very denied access.
3144 PH/28 Modified OS/os.c-Linux with
3146 -#ifndef OS_LOAD_AVERAGE
3147 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3149 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3150 nomenclature these days.)
3152 PH/29 Installed patch from the Sieve maintainer that adds the options
3153 sieve_useraddress and sieve_subaddress to the redirect router.
3155 PH/30 In these circumstances:
3156 . Two addresses routed to the same list of hosts;
3157 . First host does not offer TLS;
3158 . First host accepts first address;
3159 . First host gives temporary error to second address;
3160 . Second host offers TLS and a TLS session is established;
3161 . Second host accepts second address.
3162 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3163 and peerdn, if requested) that were in fact used only for the second
3166 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3167 attention to any local part prefix or suffix that was matched by the
3168 router that accepted the address. It now behaves in the same way as it
3169 does for delivery: the affixes are removed from the local part unless
3170 rcpt_include_affixes is set on the transport.
3172 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3173 timeout during the DATA phase of an incoming message.
3175 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3176 applied a patch sanctioned by the Sieve maintainer.
3178 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3179 the uid or gid is negative. A case of a negative gid caused this to be
3180 noticed. The fix allows for either to be negative.
3182 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3183 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3186 PH/36 The change PH/12 above was broken. Fixed it.
3188 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3189 the grounds that routing the same address twice would always produce the
3190 same answer. This might have been true once, but it is certainly no
3191 longer true now. Routing a child address may depend on the previous
3192 routing that produced that child. Some complicated redirection strategies
3193 went wrong when messages had multiple recipients, and made Exim's
3194 behaviour dependent on the order in which the addresses were given.
3196 I have moved the duplicate checking until after the routing is complete.
3197 Exim scans the addresses that are assigned to local and remote
3198 transports, and removes any duplicates. This means that more work will be
3199 done, as duplicates will always all be routed, but duplicates are
3200 presumably rare, so I don't expect this is of any significance.
3202 For deliveries to pipes, files, and autoreplies, the duplicate checking
3203 still happens during the routing process, since they are not going to be
3206 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3207 It corrects a timeout issue with spamd. This is Ian's comment: "The
3208 background is that sometimes spamd either never reads data from a
3209 connection it has accepted, or it never writes response data. The exiscan
3210 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3211 blindly assumes that writes won't block so it may never time out."
3213 PH/39 Allow G after quota size as well as K and M.
3215 PH/40 The value set for $authenticated_id in an authenticator may not contain
3216 binary zeroes or newlines because the value is written to log lines and
3217 to spool files. There was no check on this. Now the value is run through
3218 the string_printing() function so that such characters are converted to
3219 printable escape sequences.
3221 PH/41 $message_linecount is a new variable that contains the total number of
3222 lines in the message. Compare $body_linecount, which is the count for the
3225 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3226 bounce and delay warning messages, except in certain special cases, which
3229 (a) An SMTP error message from a remote host;
3230 (b) A message specified in a :fail: redirection;
3231 (c) A message specified in a "fail" command in a system filter;
3232 (d) A message specified in a FAIL return from the queryprogram router;
3233 (e) A message specified by the cannot_route_message router option.
3235 In these cases only, Exim does include the error details in bounce and
3236 warning messages. There are also a few cases where bland messages such
3237 as "unrouteable address" or "local delivery error" are given.
3239 PH/43 $value is now also set for the "else" part of a ${run expansion.
3241 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3242 being worked on, but at least Exim now implements the latest version to
3245 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3246 process to complete was treated as a delivery failure, a timeout while
3247 writing the message to the pipe was logged, but erroneously treated as a
3248 successful delivery. Such timeouts include transport filter timeouts. For
3249 consistency with the overall process timeout, these timeouts are now
3250 treated as errors, giving rise to delivery failures by default. However,
3251 there is now a new Boolean option for the pipe transport called
3252 timeout_defer, which, if set TRUE, converts the failures into defers for
3253 both kinds of timeout. A transport filter timeout is now identified in
3256 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3257 systems where "make" and "gmake" are different, calling "gmake" at top
3258 level broke things. I've arranged for the value of $(MAKE) to be passed
3259 from the Makefile to this script so that it can call the same version of
3263 A note about Exim versions 4.44 and 4.50
3264 ----------------------------------------
3266 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3267 changes of various kinds. As a consequence, a big documentation update was
3268 needed. This delayed the release for rather longer than seemed good, especially
3269 in the light of a couple of (minor) security issues. Therefore, the changes
3270 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3271 release. So 4.44 and 4.50 are in effect two different branches that both start
3274 I have left the 4.50 change log unchanged; it contains all the changes since
3275 4.43. The change log for 4.44 is below; many of its items are identical to
3276 those for 4.50. This seems to be the most sensible way to preserve the
3277 historical information.
3283 1. Minor wording change to the doc/README.SIEVE file.
3285 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3286 computation of the current number of files was incorrect.
3288 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3289 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3292 4. Give more explanation in the error message when the command for a transport
3293 filter fails to execute.
3295 5. There are several places where Exim runs a non-Exim command in a
3296 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3297 was being done only for the command run by the queryprogram router. It is
3298 now done for all such subprocesses. The other cases are: ${run, transport
3299 filters, and the commands run by the lmtp and pipe transports.
3301 6. Added CONFIGURE_GROUP build-time option.
3303 7. Some older OS have a limit of 256 on the maximum number of file
3304 descriptors. Exim was using setrlimit() to set 1000 as a large value
3305 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3306 systems. I've change it so that if it can't get 1000, it tries for 256.
3308 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3309 was an oversight, and furthermore, ever since the addition of extra
3310 controls (e.g. 4.43/32), the checks on when to allow different forms of
3311 "control" were broken. There should now be diagnostics for all cases when a
3312 control that does not make sense is encountered.
3314 9. Added the /retain_sender option to "control=submission".
3316 10. $recipients is now available in the predata ACL (oversight).
3318 11. Tidy the search cache before the fork to do a delivery from a message
3319 received from the command line. Otherwise the child will trigger a lookup
3320 failure and thereby defer the delivery if it tries to use (for example) a
3321 cached ldap connection that the parent has called unbind on.
3323 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3324 of $address_data from the recipient verification was clobbered by the
3325 sender verification.
3327 13. The value of address_data from a sender verification is now available in
3328 $sender_address_data in subsequent conditions in the ACL statement.
3330 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3332 15. Added a new option "connect=<time>" to callout options, to set a different
3335 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3336 was its contents. (It was OK if the option was not defined at all.)
3338 17. A "Completed" log line is now written for messages that are removed from
3339 the spool by the -Mrm option.
3341 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3342 information about exactly what failed.
3344 19. Added -dd to debug only the daemon process.
3346 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3347 handles timeouts, both on the server side and network timeouts. Renamed the
3348 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3350 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3351 It is now set to "smtps".
3353 22. $host_address is now set to the target address during the checking of
3354 ignore_target_hosts.
3356 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3357 being passed; this would have caused $sender_host_name to have been used if
3358 matching the list had actually called for a host name (not very likely,
3359 since this list is usually IP addresses). A host name is now passed as
3362 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3363 code that specifies a non-restarting handler (typically sigaction() in
3364 modern systems) in an attempt to fix a rare and obscure crash bug.
3366 25. Narrowed the window for a race in the daemon that could cause it to ignore
3367 SIGCHLD signals. This is not a major problem, because they are used only to
3368 wake it up if nothing else does.
3370 26. A malformed maildirsize file could cause Exim to calculate negative values
3371 for the mailbox size or file count. Odd effects could occur as a result.
3372 The maildirsize information is now recalculated if the size or filecount
3375 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3376 support for a long time. Removed HAVE_SYS_VFS_H.
3378 28. Installed the latest version of exipick from John Jetmore.
3380 29. In an address list, if the pattern was not a regular expression, an empty
3381 subject address (from a bounce message) matched only if the pattern was an
3382 empty string. Non-empty patterns were not even tested. This was the wrong
3383 because it is perfectly reasonable to use an empty address as part of a
3384 database query. An empty address is now tested by patterns that are
3385 lookups. However, all the other forms of pattern expect the subject to
3386 contain a local part and a domain, and therefore, for them, an empty
3387 address still always fails if the pattern is not itself empty.
3389 30. Exim went into a mad DNS loop when attempting to do a callout where the
3390 host was specified on an smtp transport, and looking it up yielded more
3391 than one IP address.
3393 31. Re-factored the code for checking spool and log partition space into a
3394 function that finds that data and another that does the check. The former
3395 is then used to implement four new variables: $spool_space, $log_space,
3396 $spool_inodes, and $log_inodes.
3398 32. The RFC2047 encoding function was originally intended for short strings
3399 such as real names; it was not keeping to the 75-character limit for
3400 encoded words that the RFC imposes. It now respects the limit, and
3401 generates multiple encoded words if necessary. To be on the safe side, I
3402 have increased the buffer size for the ${rfc2047: expansion operator from
3405 33. It is now permitted to omit both strings after an "if" condition; if the
3406 condition is true, the result is "true". As before, when the second string
3407 is omitted, a false condition yields an empty string. This makes it less
3408 cumbersome to write custom ACL and router conditions.
3410 34. Failure to deliver a bounce message always caused it to be frozen, even if
3411 there was an errors_to setting on the router. The errors_to setting is now
3414 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3415 canonical form (fully expanded) before being placed in
3416 $sender_host_address.
3418 36. The table in the code that translates DNS record types into text (T_A to
3419 "A" for instance) was missing entries for NS and CNAME. It is just possible
3420 that this could have caused confusion if both these types were looked up
3421 for the same domain, because the text type is used as part of Exim's
3422 per-process caching. But the chance of anyone hitting this buglet seems
3425 37. The dnsdb lookup has been extended in a number of ways.
3427 (1) There is a new type, "zns", which walks up the domain tree until it
3428 finds some nameserver records. It should be used with care.
3430 (2) There is a new type, "mxh", which is like "mx" except that it returns
3431 just the host names, not the priorities.
3433 (3) It is now possible to give a list of domains (or IP addresses) to be
3434 looked up. The behaviour when one of the lookups defers can be
3435 controlled by a keyword.
3437 (4) It is now possible to specify the separator character for use when
3438 multiple records are returned.
3440 38. The dnslists ACL condition has been extended: it is now possible to supply
3441 a list of IP addresses and/or domains to be looked up in a particular DNS
3444 39. Added log_selector=+queue_time_overall.
3446 40. When running the queue in the test harness, wait just a tad after forking a
3447 delivery process, to get repeatability of debugging output.
3449 41. Include certificate and key file names in error message when GnuTLS fails
3450 to set them up, because the GnuTLS error message doesn't include the name
3451 of the failing file when there is a problem reading it.
3453 42. Allow both -bf and -bF in the same test run.
3455 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
3457 44. The "Exiscan patch" is now merged into the mainline Exim source.
3459 45. Sometimes the final signoff response after QUIT could fail to get
3460 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3461 < 0 before doing a fflush(). This bug looks as though it goes back to the
3462 introduction of TLS in release 3.20, but "sometimes" must have been rare
3463 because the tests only now provoked it.
3465 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
3466 (this can affect the format of dates).
3468 47. exim_tidydb, when checking for the continued existence of a message for
3469 which it has found a message-specific retry record, was not finding
3470 messages that were in split spool directories. Consequently, it was
3471 deleting retry records that should have stayed in existence.
3473 48. Steve fixed some bugs in eximstats.
3475 49. The SPA authentication driver was not abandoning authentication and moving
3476 on to the next authenticator when an expansion was forced to fail,
3477 contradicting the general specification for all authenticators. Instead it
3478 was generating a temporary error. It now behaves as specified.
3480 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
3481 (the order specifies the preference for clients). The order is now AES256,
3482 AES128, 3DES, ARCFOUR128.
3484 51. Small patch to Sieve code - explicitly set From: when generating an
3487 52. Exim crashed if a remote delivery caused a very long error message to be
3488 recorded - for instance if somebody sent an entire SpamAssassin report back
3489 as a large number of 550 error lines. This bug was coincidentally fixed by
3490 increasing the size of one of Exim's internal buffers (big_buffer) that
3491 happened as part of the Exiscan merge. However, to be on the safe side, I
3492 have made the code more robust (and fixed the comments that describe what
3495 53. Now that there can be additional text after "Completed" in log lines (if
3496 the queue_time_overall log selector is set), a one-byte patch to exigrep
3497 was needed to allow it to recognize "Completed" as not the last thing in
3500 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
3501 patch that reportedly fixes this has been added. I am not expert enough to
3502 create a test for it. This is what the patch creator wrote:
3504 "I found a little strange behaviour of ldap code when working with
3505 Windows 2003 AD Domain, where users was placed in more than one
3506 Organization Units. When I tried to give exim partial DN, the exit code
3507 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
3508 But simultaneously result of request was absolutely normal ldap result,
3509 so I produce this patch..."
3511 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
3512 REFERENCE, so I have modified the code to exclude the patch when that macro
3515 55. Some experimental protocols are using DNS PTR records for new purposes. The
3516 keys for these records are domain names, not reversed IP addresses. The
3517 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3518 leaves it alone. Component reversal etc. now happens only for IP addresses.
3521 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
3523 57. Double the size of the debug message buffer (to 2048) so that more of very
3524 long debug lines gets shown.
3526 58. The exicyclog utility now does better if the number of log files to keep
3527 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
3529 59. Two changes related to the smtp_active_hostname option:
3531 (1) $smtp_active_hostname is now available as a variable.
3532 (2) The default for smtp_banner uses $smtp_active_hostname instead
3533 of $primary_hostname.
3535 60. The host_aton() function is supposed to be passed a string that is known
3536 to be a valid IP address. However, in the case of IPv6 addresses, it was
3537 not checking this. This is a hostage to fortune. Exim now panics and dies
3538 if the condition is not met. A case was found where this could be provoked
3539 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3540 components; fortuitously, this particular loophole had already been fixed
3541 by change 4.50/55 above.
3543 If there are any other similar loopholes, the new check in host_aton()
3544 itself should stop them being exploited. The report I received stated that
3545 data on the command line could provoke the exploit when Exim was running as
3546 exim, but did not say which command line option was involved. All I could
3547 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3548 running as the user.
3551 61. There was a buffer overflow vulnerability in the SPA authentication code
3552 (which came originally from the Samba project). I have added a test to the
3553 spa_base64_to_bits() function which I hope fixes it.
3556 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
3557 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
3559 63. The daemon start-up calls getloadavg() while still root for those OS that
3560 need the first call to be done as root, but it missed one case: when
3561 deliver_queue_load_max is set with deliver_drop_privilege. This is
3562 necessary for the benefit of the queue runner, because there is no re-exec
3563 when deliver_drop_privilege is set.
3565 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
3566 This has been fixed.
3568 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3569 was in use, was not putting the data itself into the right store pool;
3570 consequently, it could be overwritten for a subsequent message in the same
3571 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3574 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
3576 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
3577 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
3578 the function were treating the return as a boolean value, which happened to
3579 work because 0=false and not-0=true, but is not correct code.
3581 68. The host_aton() function was not handling scoped IPv6 addresses (those
3582 with, for example, "%eth0" on the end) correctly.
3584 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
3585 negated items (that is, ~something) in unsigned ints. Some compilers
3586 apparently mutter when there is no cast.
3588 70. If an address verification called from an ACL failed, and did not produce a
3589 user-specific message (i.e. there was only a "system" message), nothing was
3590 put in $acl_verify_message. In this situation, it now puts the system
3593 71. Change 4.23/11 added synchronization checking at the start of an SMTP
3594 session; change 4.31/43 added the unwanted input to the log line - except
3595 that it did not do this in the start of session case. It now does.
3597 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
3598 This is wrong and can cause the other end to generate a synchronization
3599 error if it is another Exim or anything else that does the synchronization
3600 check. A QUIT command is no longer sent after a timeout.
3602 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
3603 during host lookups.
3605 74. The defer_ok option of callout verification was not working if it was used
3606 when verifying addresses in header lines, that is, for this case:
3608 verify = header_sender/callout=defer_ok
3610 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
3611 those file descriptors could be used for SMTP connections. If anything
3612 wrote to stderr (the example that came up was "warn" in embedded Perl), it
3613 could be sent to the SMTP client, causing chaos. The daemon now opens
3614 stdin, stdout, and stderr to /dev/null when it puts itself into the
3617 76. Arrange for output from Perl's "warn" command to be written to Exim's main
3618 log by default. The user can override this with suitable Perl magic.
3620 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
3621 the log message when discard triggers, was not working for the DATA ACL or
3622 for the non-SMTP ACL.
3624 78. Error message wording change in sieve.c.
3626 79. If smtp_accept_max_per_host was set, the number of connections could be
3627 restricted to fewer than expected, because the daemon was trying to set up
3628 a new connection before checking whether the processes handling previous
3629 connections had finished. The check for completed processes is now done
3630 earlier. On busy systems, this bug wouldn't be noticed because something
3631 else would have woken the daemon, and it would have reaped the completed
3634 80. If a message was submitted locally by a user whose login name contained one
3635 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
3636 It caused a spool format error. I have fixed the spool reading code. A
3637 related problem was that the "from" clause in the Received: line became
3638 illegal because of the space(s). It is now covered by ${quote_local_part.
3640 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
3643 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
3645 83. Patch from Sieve maintainer to fix unterminated string problem in
3646 "vacation" handling.
3648 84. Some minor changes to the Linux configuration files to help with other
3649 OS variants using glibc.
3651 85. One more patch for Sieve to update vacation handling to latest spec.
3654 ----------------------------------------------------
3655 See the note above about the 4.44 and 4.50 releases.
3656 ----------------------------------------------------
3662 1. Change 4.43/35 introduced a bug that caused file counts to be
3663 incorrectly computed when quota_filecount was set in an appendfile
3666 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3667 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3670 3. Give more explanation in the error message when the command for a transport
3671 filter fails to execute.
3673 4. There are several places where Exim runs a non-Exim command in a
3674 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3675 was being done only for the command run by the queryprogram router. It is
3676 now done for all such subprocesses. The other cases are: ${run, transport
3677 filters, and the commands run by the lmtp and pipe transports.
3679 5. Some older OS have a limit of 256 on the maximum number of file
3680 descriptors. Exim was using setrlimit() to set 1000 as a large value
3681 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3682 systems. I've change it so that if it can't get 1000, it tries for 256.
3684 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3685 was an oversight, and furthermore, ever since the addition of extra
3686 controls (e.g. 4.43/32), the checks on when to allow different forms of
3687 "control" were broken. There should now be diagnostics for all cases when a
3688 control that does not make sense is encountered.
3690 7. $recipients is now available in the predata ACL (oversight).
3692 8. Tidy the search cache before the fork to do a delivery from a message
3693 received from the command line. Otherwise the child will trigger a lookup
3694 failure and thereby defer the delivery if it tries to use (for example) a
3695 cached ldap connection that the parent has called unbind on.
3697 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3698 of $address_data from the recipient verification was clobbered by the
3699 sender verification.
3701 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3702 was its contents. (It was OK if the option was not defined at all.)
3704 11. A "Completed" log line is now written for messages that are removed from
3705 the spool by the -Mrm option.
3707 12. $host_address is now set to the target address during the checking of
3708 ignore_target_hosts.
3710 13. When checking ignore_target_hosts for an ipliteral router, no host name was
3711 being passed; this would have caused $sender_host_name to have been used if
3712 matching the list had actually called for a host name (not very likely,
3713 since this list is usually IP addresses). A host name is now passed as
3716 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3717 code that specifies a non-restarting handler (typically sigaction() in
3718 modern systems) in an attempt to fix a rare and obscure crash bug.
3720 15. Narrowed the window for a race in the daemon that could cause it to ignore
3721 SIGCHLD signals. This is not a major problem, because they are used only to
3722 wake it up if nothing else does.
3724 16. A malformed maildirsize file could cause Exim to calculate negative values
3725 for the mailbox size or file count. Odd effects could occur as a result.
3726 The maildirsize information is now recalculated if the size or filecount
3729 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3730 support for a long time. Removed HAVE_SYS_VFS_H.
3732 18. Updated exipick to current release from John Jetmore.
3734 19. Allow an empty sender to be matched against a lookup in an address list.
3735 Previously the only cases considered were a regular expression, or an
3738 20. Exim went into a mad DNS lookup loop when doing a callout where the
3739 host was specified on the transport, if the DNS lookup yielded more than
3742 21. The RFC2047 encoding function was originally intended for short strings
3743 such as real names; it was not keeping to the 75-character limit for
3744 encoded words that the RFC imposes. It now respects the limit, and
3745 generates multiple encoded words if necessary. To be on the safe side, I
3746 have increased the buffer size for the ${rfc2047: expansion operator from
3749 22. Failure to deliver a bounce message always caused it to be frozen, even if
3750 there was an errors_to setting on the router. The errors_to setting is now
3753 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3754 canonical form (fully expanded) before being placed in
3755 $sender_host_address.
3757 24. Updated eximstats to version 1.33
3759 25. Include certificate and key file names in error message when GnuTLS fails
3760 to set them up, because the GnuTLS error message doesn't include the name
3761 of the failing file when there is a problem reading it.
3763 26. Expand error message when OpenSSL has problems setting up cert/key files.
3766 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
3767 (this can affect the format of dates).
3769 28. exim_tidydb, when checking for the continued existence of a message for
3770 which it has found a message-specific retry record, was not finding
3771 messages that were in split spool directories. Consequently, it was
3772 deleting retry records that should have stayed in existence.
3774 29. eximstats updated to version 1.35
3775 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
3776 1.35 - bugfix such that pie charts by volume are generated correctly
3778 30. The SPA authentication driver was not abandoning authentication and moving
3779 on to the next authenticator when an expansion was forced to fail,
3780 contradicting the general specification for all authenticators. Instead it
3781 was generating a temporary error. It now behaves as specified.
3783 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
3784 (the order specifies the preference for clients). The order is now AES256,
3785 AES128, 3DES, ARCFOUR128.
3787 31. Small patch to Sieve code - explicitly set From: when generating an
3790 32. Exim crashed if a remote delivery caused a very long error message to be
3791 recorded - for instance if somebody sent an entire SpamAssassin report back
3792 as a large number of 550 error lines. This bug was coincidentally fixed by
3793 increasing the size of one of Exim's internal buffers (big_buffer) that
3794 happened as part of the Exiscan merge. However, to be on the safe side, I
3795 have made the code more robust (and fixed the comments that describe what
3798 33. Some experimental protocols are using DNS PTR records for new purposes. The
3799 keys for these records are domain names, not reversed IP addresses. The
3800 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3801 leaves it alone. Component reversal etc. now happens only for IP addresses.
3804 34. The host_aton() function is supposed to be passed a string that is known
3805 to be a valid IP address. However, in the case of IPv6 addresses, it was
3806 not checking this. This is a hostage to fortune. Exim now panics and dies
3807 if the condition is not met. A case was found where this could be provoked
3808 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3809 components; fortuitously, this particular loophole had already been fixed
3810 by change 4.50/55 or 4.44/33 above.
3812 If there are any other similar loopholes, the new check in host_aton()
3813 itself should stop them being exploited. The report I received stated that
3814 data on the command line could provoke the exploit when Exim was running as
3815 exim, but did not say which command line option was involved. All I could
3816 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3817 running as the user.
3820 35. There was a buffer overflow vulnerability in the SPA authentication code
3821 (which came originally from the Samba project). I have added a test to the
3822 spa_base64_to_bits() function which I hope fixes it.
3825 36. The daemon start-up calls getloadavg() while still root for those OS that
3826 need the first call to be done as root, but it missed one case: when
3827 deliver_queue_load_max is set with deliver_drop_privilege. This is
3828 necessary for the benefit of the queue runner, because there is no re-exec
3829 when deliver_drop_privilege is set.
3831 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3832 was in use, was not putting the data itself into the right store pool;
3833 consequently, it could be overwritten for a subsequent message in the same
3834 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3837 38. Sometimes the final signoff response after QUIT could fail to get
3838 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3839 < 0 before doing a fflush(). This bug looks as though it goes back to the
3840 introduction of TLS in release 3.20, but "sometimes" must have been rare
3841 because the tests only now provoked it.
3847 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
3848 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
3849 Now it returns an integer. A number of calls were still expecting a T/F
3850 return. Fortuitously, in all cases, the tests worked in OK situations,
3851 which is the norm. However, things would have gone wrong on any write
3852 failures on the smtp file descriptor. This function is used when sending
3853 messages over SMTP and also when doing verify callouts.
3855 2. When Exim is called to do synchronous delivery of a locally submitted
3856 message (the -odf or -odi options), it no longer closes stderr before doing
3859 3. Implemented the mua_wrapper option.
3861 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
3863 5. Implemented the functions header_remove(), header_testname(),
3864 header_add_at_position(), and receive_remove_recipient(), and exported them
3867 6. If an ACL "warn" statement specified the addition of headers, Exim already
3868 inserted X-ACL-Warn: at the start if there was no header name. However, it
3869 was not making this test for the second and subsequent header lines if
3870 there were newlines in the string. This meant that an invalid header could
3871 be inserted if Exim was badly configured.
3873 7. Allow an ACL "warn" statement to add header lines at the start or after all
3874 the Received: headers, as well as at the end.
3876 8. Added the rcpt_4xx retry error code.
3878 9. Added postmaster_mailfrom=xxx to callout verification option.
3880 10. Added mailfrom=xxxx to the callout verification option, for verify=
3883 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
3884 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
3886 12. Inserted some casts to stop certain compilers warning when using pointer
3887 differences as field lengths or precisions in printf-type calls (mostly
3888 affecting debugging statements).
3890 13. Added optional readline() support for -be (dynamically loaded).
3892 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
3893 same clock tick as a message's arrival, so that its received time was the
3894 same as the "first fail" time on the retry record, and that message
3895 remained on the queue past the ultimate address timeout, every queue runner
3896 would try a delivery (because it was past the ultimate address timeout) but
3897 after another failure, the ultimate address timeout, which should have then
3898 bounced the address, did not kick in. This was a "< instead of <=" error;
3899 in most cases the first failure would have been in the next clock tick
3900 after the received time, and all would be well.
3902 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
3903 being recognized when the domain list was tested by the match_domain
3904 condition in an expansion string.
3906 16. Added the ${str2b64: operator.
3908 17. Exim was always calling setrlimit() to set a large limit for the number of
3909 processes, without checking whether the existing limit was already
3910 adequate. (It did check for the limit on file descriptors.) Furthermore,
3911 errors from getrlimit() and setrlimit() were being ignored. Now they are
3912 logged to the main and panic logs, but Exim does carry on, to try to do its
3913 job under whatever limits there are.
3915 18. Imported PCRE 5.0.
3917 19. Trivial typo in log message " temporarily refused connection" (the leading
3920 20. If the log selector return_path_on_delivery was set and an address was
3921 redirected to /dev/null, the delivery process crashed because it assumed
3922 that a return path would always be set for a "successful" delivery. In this
3923 case, the whole delivery is bypassed as an optimization, and therefore no
3926 21. Internal re-arrangement: the function for sending a challenge and reading
3927 a response while authentication was assuming a zero-terminated challenge
3928 string. It's now changed to take a pointer and a length, to allow for
3929 binary data in such strings.
3931 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3933 23. Exim was not respecting finduser_retries when seeking the login of the
3934 uid under which it was called; it was always trying 10 times. (The default
3935 setting of finduser_retries is zero.) Also, it was sleeping after the final
3936 failure, which is pointless.
3938 24. Implemented tls_on_connect_ports.
3940 25. Implemented acl_smtp_predata.
3942 26. If the domain in control=submission is set empty, Exim assumes that the
3943 authenticated id is a complete email address when it generates From: or
3944 Sender: header lines.
3946 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3947 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3948 chown and chgrp in /bin and hostname in /usr/bin.
3950 28. Exim was keeping the "process log" file open after each use, just as it
3951 does for the main log. This opens the possibility of it remaining open for
3952 long periods when the USR1 signal hits a daemon. Occasional processlog
3953 errors were reported, that could have been caused by this. Anyway, it seems
3954 much more sensible not to leave this file open at all, so that is what now
3957 29. The long-running daemon process does not normally write to the log once it
3958 has entered its main loop, and it closes the log before doing so. This is
3959 so that log files can straightforwardly be renamed and moved. However,
3960 there are a couple of unusual error situations where the daemon does write
3961 log entries, and I had neglected to close the log afterwards.
3963 30. The text of an SMTP error response that was received during a remote
3964 delivery was being truncated at 512 bytes. This is too short for some of
3965 the long messages that one sometimes sees. I've increased the limit to
3968 31. It is now possible to make retry rules that apply only when a message has a
3969 specific sender, in particular, an empty sender.
3971 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3972 it possible to be selective about when SMTP synchronization is enforced.
3974 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3976 32. Implemented hosts_connection_nolog.
3978 33. Added an ACL for QUIT.
3980 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3983 35. Added mailbox_size and mailbox_filecount to appendfile.
3985 36. Added control = no_multiline_responses to ACLs.
3987 37. There was a bug in the logic of the code that waits for the clock to tick
3988 in the case where the clock went backwards by a substantial amount such
3989 that the microsecond fraction of "now" was more than the microsecond
3990 fraction of "then" (but the whole seconds number was less).
3992 38. Added support for the libradius Radius client library this is found on
3993 FreeBSD (previously only the radiusclient library was supported).
3999 1. When certain lookups returned multiple values in the form name=value, the
4000 quoting of the values was not always being done properly. Specifically:
4001 (a) If the value started with a double quote, but contained no whitespace,
4003 (b) If the value contained whitespace other than a space character (i.e.
4004 tabs or newlines or carriage returns) it was not quoted.
4005 This fix has been applied to the mysql and pgsql lookups by writing a
4006 separate quoting function and calling it from the lookup code. The fix
4007 should probably also be applied to nisplus, ibase and oracle lookups, but
4008 since I cannot test any of those, I have not disturbed their existing code.
4010 2. A hit in the callout cache for a specific address caused a log line with no
4011 reason for rejecting RCPT. Now it says "Previous (cached) callout
4012 verification failure".
4014 3. There was an off-by-one bug in the queryprogram router. An over-long
4015 return line was truncated at 256 instead of 255 characters, thereby
4016 overflowing its buffer with the terminating zero. As well as fixing this, I
4017 have increased the buffer size to 1024 (and made a note to document this).
4019 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4020 when Exim is waiting for an SMTP response from a remote server, Exim
4021 restarts its select() call on the socket, thereby resetting its timeout.
4022 This is not a problem when such interrupts are rare. Somebody set up a cron
4023 job to run exiwhat every 2 minutes, which is less than the normal select()
4024 timeout (5 or 10 minutes). This meant that the select() timeout never
4025 kicked in because it was always reset. I have fixed this by comparing the
4026 time when an interrupt arrives with the time at the start of the first call
4027 to select(). If more time than the timeout has elapsed, the interrupt is
4028 treated as a timeout.
4030 5. Some internal re-factoring in preparation for the addition of Sieve
4031 extensions (by MH). In particular, the "personal" test is moved to a
4032 separate function, and given an option for scanning Cc: and Bcc: (which is
4033 not set for Exim filters).
4035 6. When Exim created an email address using the login of the caller as the
4036 local part (e.g. when creating a From: or Sender: header line), it was not
4037 quoting the local part when it contained special characters such as @.
4039 7. Installed new OpenBSD configuration files.
4041 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4042 try to make them clearer.
4044 9. Callout options, other than the timeout value, were being ignored when
4045 verifying sender addresses in header lines. For example, when using
4047 verify = header_sender/callout=no_cache
4049 the cache was (incorrectly) being used.
4051 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4052 only the Cygwin environment.
4054 11. When return_path_on_delivery was set as a log selector, if different remote
4055 addresses in the same message used different return paths and parallel
4056 remote delivery occurred, the wrong values would sometimes be logged.
4057 (Whenever a remote delivery process finished, the return path value from
4058 the most recently started remote delivery process was logged.)
4060 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4061 lines when AUTH and/or TLS are in use. This is the "received protocol"
4062 field. Exim used to use "asmtp" for authenticated SMTP, without any
4063 indication (in the protocol name) for TLS use. Now it follows the RFC and
4064 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4065 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4066 names appear in log lines as well as in Received: header lines.
4068 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4069 extensions, and comparison tests, and to fix some bugs.
4071 14. Changes to the "personal" filter test:
4073 (1) The test was buggy in that it was just doing the equivalent of
4074 "contains" tests on header lines. For example, if a user's address was
4075 anne@some.where, the "personal" test would incorrectly be true for
4077 To: susanne@some.where
4079 This test is now done by extracting each address from the header in turn,
4080 and checking the entire address. Other tests that are part of "personal"
4081 are now done using regular expressions (for example, to check local parts
4082 of addresses in From: header lines).
4084 (2) The list of non-personal local parts in From: addresses has been
4085 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4086 taken from the Sieve specification recommendations.
4088 (3) If the message contains any header line starting with "List-" it is
4089 treated as non-personal.
4091 (4) The test for "circular" in the Subject: header line has been removed
4092 because it now seems ill-conceived.
4094 15. Minor typos in src/EDITME comments corrected.
4096 16. Installed latest exipick from John Jetmore.
4098 17. If headers_add on a router specified a text string that was too long for
4099 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4100 of string_sprintf() is now avoided.
4102 18. $message_body_size was not set (it was always zero) when running the DATA
4103 ACL and the local_scan() function.
4105 19. For the "mail" command in an Exim filter, no default was being set for
4106 the once_repeat time, causing a random time value to be used if "once" was
4107 specified. (If the value happened to be <= 0, no repeat happened.) The
4108 default is now 0s, meaning "never repeat". The "vacation" command was OK
4109 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4110 (I found it when inspecting the code).
4112 20. There is now an overall timeout for performing a callout verification. It
4113 defaults to 4 times the callout timeout, which applies to individual SMTP
4114 commands during the callout. The overall timeout applies when there is more
4115 than one host that can be tried. The timeout is checked before trying the
4116 next host. This prevents very long delays if there are a large number of
4117 hosts and all are timing out (e.g. when the network connections are timing
4118 out). The value of the overall timeout can be changed by specifying an
4119 additional sub-option for "callout", called "maxwait". For example:
4121 verify = sender/callout=5s,maxwait=20s
4123 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4124 to the end before writing, but this should make it even safer).
4126 22. Exim was forgetting that it had advertised PIPELINING for the second and
4127 subsequent messages on an SMTP connection. It was also not resetting its
4128 memory on STARTTLS and an internal HELO.
4130 23. When Exim logs an SMTP synchronization error within a session, it now
4131 records whether PIPELINING has been advertised or not.
4133 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4134 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4135 rather than long int.
4137 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4139 26. Added the never_mail option to autoreply.
4145 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4146 crash if the getsockname() call failed; this can happen if a connection is
4147 closed very soon after it is established. The problem was simply in the
4148 order in which certain operations were done, causing Exim to try to write
4149 to the SMTP stream before it had set up the file descriptor. The bug has
4150 been fixed by making things happen in the correct order.
4156 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4157 before the connection was closed, thus losing the rejection response.
4159 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4160 some early Solaris releases, but causes trouble in current releases where
4161 socklen_t is defined.
4163 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4166 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4169 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4170 that releasing the top of it at the end releases what was used for sub-
4171 expansions (unless the block got too big). However, discard this block if
4172 the first thing is a variable or header, so that we can use its block when
4173 it is dynamic (useful for very large $message_headers, for example).
4175 6. Lookups now cache *every* query, not just the most recent. A new, separate
4176 store pool is used for this. It can be recovered when all lookup caches are
4177 flushed. Lookups now release memory at the end of their result strings.
4178 This has involved some general refactoring of the lookup sources.
4180 7. Some code has been added to the store_xxx() functions to reduce the amount
4181 of flapping under certain conditions.
4183 8. log_incoming_interface used to affect only the <= reception log lines. Now
4184 it causes the local interface and port to be added to several more SMTP log
4185 lines, for example "SMTP connection from", and rejection lines.
4187 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4189 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4191 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4192 could be overwritten at the end of the current message (or the start of a
4193 new message if it was set in a HELO ACL). The value is now preserved for
4194 the duration of the SMTP connection.
4196 12. If a transport had a headers_rewrite setting, and a matching header line
4197 contained an unqualified address, that address was qualified, even if it
4198 did not match any rewriting rules. The underlying bug was that the values
4199 of the flags that permit the existence of unqualified sender and recipient
4200 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4201 non-local messages, and by -bnq for local messages) were not being
4202 preserved with the message after it was received.
4204 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4205 "next input=" as part of the text comprising the host identity instead of
4206 the correct text. The code was using the same buffer for two different
4207 strings. However, depending on which order the printing function evaluated
4208 its arguments, the bug did not always show up. Under Linux, for example, my
4209 test suite worked just fine.
4211 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4212 doesn't work with some older versions of Perl. It has been changed to "my",
4213 which in any case is probably the better facility to use.
4215 15. A really picky compiler found some instances of statements for creating
4216 error messages that either had too many or two few arguments for the format
4219 16. The size of the buffer for calls to the DNS resolver has been increased
4220 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4221 for addresses that have a lot of PTR records. This alleviates a problem; it
4222 does not fully solve it.
4224 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4225 buffer now truncates the list and logs the incident, which is the same
4226 action as happens when Exim is looking up a host name and its aliases.
4227 Previously in this situation something unpredictable would happen;
4228 sometimes it was "internal error: store_reset failed".
4230 18. If a server dropped the connection unexpectedly when an Exim client was
4231 using GnuTLS and trying to read a response, the client delivery process
4232 crashed while trying to generate an error log message.
4234 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4235 string, for example:
4237 warn message = H1: something\nH2: something
4239 the text was added as a single header line from Exim's point of view
4240 though it ended up OK in the delivered message. However, searching for the
4241 second and subsequent header lines using $h_h2: did not work. This has been
4242 fixed. Similarly, if a system filter added multiple headers in this way,
4243 the routers could not see them.
4245 20. Expanded the error message when iplsearch is called with an invalid key to
4246 suggest using net-iplsearch in a host list.
4248 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4249 ACLs are no longer actually imposed (and a message to that effect is
4252 22. If a "gecos" field in a passwd entry contained escaped characters, in
4253 particular, if it contained a \" sequence, Exim got it wrong when building
4254 a From: or a Sender: header from that name. A second bug also caused
4255 incorrect handling when an unquoted " was present following a character
4256 that needed quoting.
4258 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4259 was not being matched caselessly.
4261 24. Arranged for all hyphens in the exim.8 source to be escaped with
4264 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4265 properly. Recipient callout cache records were still being keyed to include
4266 the sender, even when use_sender was set false. This led to far more
4267 callouts that were necessary. The sender is no longer included in the key
4268 when use_sender is false.
4270 26. Added "control = submission" modifier to ACLs.
4272 27. Added the ${base62d: operator to decode base 62 numbers.
4274 28. dnsdb lookups can now access SRV records.
4276 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4277 the configuration file.
4279 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4280 (-v) mode. This makes the output for a verbose queue run more intelligible.
4282 31. Added a use_postmaster feature to recipient callouts.
4284 32. Added the $body_zerocount variable, containing the number of binary zero
4285 bytes in the message body.
4287 33. The time of last modification of the "new" subdirectory is now used as the
4288 "mailbox time last read" when there is a quota error for a maildir
4291 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4293 35. Added +ignore_unknown as a special item in host lists.
4295 36. Code for decoding IPv6 addresses in host lists is now included, even if
4296 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4297 address was recognized as an IP address, but was then not correctly decoded
4298 into binary, causing unexpected and incorrect effects when compared with
4305 1. Very minor rewording of debugging text in manualroute to say "list of
4306 hosts" instead of "hostlist".
4308 2. If verify=header_syntax was set, and a header line with an unqualified
4309 address (no domain) and a large number of spaces between the end of the
4310 name and the colon was received, the reception process suffered a buffer
4311 overflow, and (when I tested it) crashed. This was caused by some obsolete
4312 code that should have been removed. The fix is to remove it!
4314 3. When running in the test harness, delay a bit after writing a bounce
4315 message to get a bit more predictability in the log output.
4317 4. Added a call to search_tidyup() just before forking a reception process. In
4318 theory, someone could use a lookup in the expansion of smtp_accept_max_
4319 per_host which, without the tidyup, could leave open a database connection.
4321 5. Added the variables $recipient_data and $sender_data which get set from a
4322 lookup success in an ACL "recipients" or "senders" condition, or a router
4323 "senders" option, similar to $domain_data and $local_part_data.
4325 6. Moved the writing of debug_print from before to after the "senders" test
4328 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4329 problems for message scanning, either using a data ACL, or using
4330 local_scan() because the Received: header was not generated till after they
4331 were called (in order to set the time as the time of reception completion).
4332 I have revised the way this works. The header is now generated after the
4333 body is received, but before the ACL or local_scan() are called. After they
4334 are run, the timestamp in the header is updated.
4340 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4341 before starting a queue runner without re-exec. This happened only when
4342 deliver_drop_privilege was set or when the Exim user was set to root. The
4343 effect of the bug was that timeouts during subsequent deliveries caused
4344 crashes instead of being properly handled. The handler is now left at its
4345 default (and expected) setting.
4347 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4348 message, again when deliver_drop_privilege is set or Exim is run as root.
4349 The bug described in (1) was not present in this case, but the tidying up
4350 of the other signals was missing. I have made the two cases consistent.
4352 3. The ignore_target_hosts setting on a manualroute router was being ignored
4353 for hosts that were looked up using the /MX notation.
4355 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4358 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4359 operated on the sender address. After changing the $sender_address to <>
4360 for the sender address verify, Exim was re-instated it as the original
4361 (before rewriting) address, but remembering that it had rewritten it, so it
4362 wasn't rewriting it again. This bug also had the effect of breaking the
4363 sender address verification caching when the sender address was rewritten.
4365 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4366 This has been changed so that if the ip literal address matches
4367 ignore_target_hosts, the router declines.
4369 7. Added expansion conditions match_domain, match_address, and match_local_
4370 part (NOT match_host).
4372 8. The placeholder for the Received: header didn't have a length field set.
4374 9. Added code to Exim itself and to exim_lock to test for a specific race
4375 condition that could lead to file corruption when using MBX delivery. The
4376 issue is with the lockfile that is created in /tmp. If this file is removed
4377 after a process has opened it but before that process has acquired a lock,
4378 there is the potential for a second process to recreate the file and also
4379 acquire a lock. This could lead to two Exim processes writing to the file
4380 at the same time. The added code performs the same test as UW imapd; it
4381 checks after acquiring the lock that its file descriptor still refers to
4382 the same named file.
4384 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4385 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4388 11. Added the smtp_active_hostname option. If used, this will typically be made
4389 to depend on the incoming interface address. Because $interface_address is
4390 not set up until the daemon has forked a reception process, error responses
4391 that can happen earlier (such as "too many connections") no longer contain
4394 12. If an expansion in a condition on a "warn" statement fails because a lookup
4395 defers, the "warn" statement is abandoned, and the next ACL statement is
4396 processed. Previously this caused the whole ACL to be aborted.
4398 13. Added the iplsearch lookup type.
4400 14. Added ident_timeout as a log selector.
4402 15. Added tls_certificate_verified as a log selector.
4404 16. Added a global option tls_require_ciphers (compare the smtp transport
4405 option of the same name). This controls incoming TLS connections.
4407 17. I finally figured out how to make tls_require_ciphers do a similar thing
4408 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
4409 before starting the TLS session.
4411 18. Tabs are now shown as \t in -bP output.
4413 19. If the log selector return_path_on_delivery was set, Exim crashed when
4414 bouncing a message because it had too many Received: header lines.
4416 20. If two routers both had headers_remove settings, and the first one included
4417 a superfluous trailing colon, the final name in the first list and the
4418 first name in the second list were incorrectly joined into one item (with a
4419 colon in the middle).
4425 1. Added -C and -D options to the exinext utility, mainly to make it easier
4426 to include in the automated testing, but these could be helpful when
4427 multiple configurations are in use.
4429 2. The exinext utility was not formatting the output nicely when there was
4430 an alternate port involved in the retry record key, nor when there was a
4431 message id as well (for retries that were specific to a specific message
4432 and a specific host). It was also confused by IPv6 addresses, because of
4433 the additional colons they contain. I have fixed the IPv4 problem, and
4434 patched it up to do a reasonable job for IPv6.
4436 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
4437 delivery, the log line now contains "pipelined" if PIPELINING was used.
4439 4. An SMTP transport process used to panic and die if the bind() call to set
4440 an explicit outgoing interface failed. This has been changed; it is now
4441 treated in the same way as a connect() failure.
4443 5. A reference to $sender_host_name in the part of a conditional expansion
4444 that was being skipped was still causing a DNS lookup. This no longer
4447 6. The def: expansion condition was not recognizing references to header lines
4448 that used bh_ and bheader_.
4450 7. Added the _cache feature to named lists.
4452 8. The code for checking quota_filecount in the appendfile transport was
4453 allowing one more file than it should have been.
4455 9. For compatibility with Sendmail, the command line option
4463 and sets the incoming protocol and host name (for trusted callers). The
4464 host name and its colon can be omitted when only the protocol is to be set.
4465 Note the Exim already has two private options, -pd and -ps, that refer to
4466 embedded Perl. It is therefore impossible to set a protocol value of "d" or
4467 "s", but I don't think that's a major issue.
4469 10. A number of refactoring changes to the code, none of which should affect
4472 (a) The number of logging options was getting close to filling up the
4473 32-bit word that was used as a bit map. I have split them into two classes:
4474 those that are passed in the argument to log_write(), and those that are
4475 only ever tested independently outside of that function. These are now in
4476 separate 32-bit words, so there is plenty of room for expansion again.
4477 There is no change in the user interface or the logging behaviour.
4479 (b) When building, for example, log lines, the code previously used a
4480 macro that called string_cat() twice, in order to add two strings. This is
4481 not really sufficiently general. Furthermore, there was one instance where
4482 it was actually wrong because one of the argument was used twice, and in
4483 one call a function was used. (As it happened, calling the function twice
4484 did not affect the overall behaviour.) The macro has been replaced by a
4485 function that can join an arbitrary number of extra strings onto a growing
4488 (c) The code for expansion conditions now uses a table and a binary chop
4489 instead of a serial search (which was left over from when there were very
4490 few conditions). Also, it now recognizes conditions like "pam" even when
4491 the relevant support is not compiled in: a suitably worded error message is
4492 given if an attempt is made to use such a condition.
4494 11. Added ${time_interval:xxxxx}.
4496 12. A bug was causing one of the ddress fields not to be passed back correctly
4497 from remote delivery subprocesses. The field in question was not being
4498 subsequently used, so this caused to problems in practice.
4500 13. Added new log selectors queue_time and deliver_time.
4502 14. Might have fixed a bug in maildirsizefile handling that threw up
4503 "unexpected character" debug warnings, and recalculated the data
4504 unnecessarily. In any case, I expanded the warning message to give more
4507 15. Added the message "Restricted characters in address" to the statements in
4508 the default ACL that block characters like @ and % in local parts.
4510 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
4511 Three changes have been made:
4513 (a) There was a serious bug; a negative response to MAIL caused the whole
4514 recipient domain to be cached as invalid, thereby blocking all messages
4515 to all local parts at the same domain, from all senders. This bug has
4516 been fixed. The domain is no longer cached after a negative response to
4517 MAIL if the sender used is not empty.
4519 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
4522 (c) A new callout option, "use_sender" has been added for people who want
4523 the modified behaviour.
4529 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
4532 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
4533 indeed breaks things for older releases.
4535 3. Added additional logging to the case where there is a problem reading data
4536 from a filter that is running in a subprocess using a pipe, in order to
4537 try to track down a specific problem.
4539 4. Testing facility fudge: when running in the test harness and attempting
4540 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
4541 getting "No route to host". Convert this to a timeout.
4543 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
4546 6. Some OS don't have socklen_t but use size_t instead. This affects the
4547 fifth argument of getsockopt() amongst other things. This is now
4548 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
4549 can be set for individual OS. I have set it for SunOS5, OSF1, and
4550 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
4551 some earlier ones do not.
4553 7. Change 4.30/15 was not doing the test caselessly.
4555 8. The standard form for an IPv6 address literal was being rejected by address
4556 parsing in, for example, MAIL and RCPT commands. An example of this kind of
4557 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4558 this, as well as the form without the "IPv6" on the front (but only when
4559 address literals are enabled, of course).
4561 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
4563 10. Exim crashed if a message with an empty sender address specified by -f
4564 encountered a router with an errors_to setting. This could be provoked only
4565 by a command such as
4569 where an empty string was supplied; "<>" did not hit this bug.
4571 11. Installed PCRE release 4.5.
4573 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
4574 remained set. It is now erased.
4576 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
4577 times from message ids (which are base 36 rather than the normal 62).
4579 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
4580 were being counted as actual protocol errors, and logged if the log
4581 selector +smtp_protocol_error was set. One cannot be perfect in this test,
4582 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
4583 and DATA following a set of rejected RCPTs do not count as protocol errors.
4584 In other words, Exim assumes they were pipelined, though this may not
4585 actually be the case. Of course, in all cases the client gets an
4586 appropriate error code.
4588 15. If a lookup fails in an ACL condition, a message about the failure may
4589 be available; it is used if testing the ACL cannot continue, because most
4590 such messages specify what the cause of the deferral is. However, some
4591 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
4592 that caused an old message to be retained and used if a later statement
4593 caused a defer, replacing the real cause of the deferral.
4595 16. If an IP address had so many PTR records that the DNS lookup buffer
4596 was not large enough to hold them, Exim could crash while trying to process
4597 the truncated data. It now detects and logs this case.
4599 17. Further to 4.21/58, another change has been made: if (and only if) the
4600 first line of a message (the first header line) ends with CRLF, a bare LF
4601 in a subsequent header line has a space inserted after it, so as not to
4602 terminate the header.
4604 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
4605 unnecessarily, used atoi() instead of strtol(), and didn't check the
4606 termination when getting file sizes from file names by regex.
4608 19. Completely re-implemented the support for maildirsize files, in the light
4609 of a number of problems with the previous contributed implementation
4610 (4.30/29). In particular:
4612 . If the quota is zero, the maildirsize file is maintained, but no quota is
4615 . If the maildir directory does not exist, it is created before any attempt
4616 to write a maildirsize file.
4618 . The quota value in the file is just a cache; if the quota is changed in
4619 the transport, the new value overrides.
4621 . A regular expression is available for excluding directories from the
4624 20. The autoreply transport checks the characters in options that define the
4625 message's headers; it allows continued headers, but it was checking with
4626 isspace() after an embedded newline instead of explicitly looking for a
4629 21. If all the "regular" hosts to which an address was routed had passed their
4630 expiry times, and had not reached their retry times, the address was
4631 bounced, even if fallback hosts were defined. Now Exim should go on to try
4634 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
4635 equivalent code in the SMTP transport. Some hosts send humungous responses
4636 to HELO/EHLO, more than 1024 it seems.
4638 23. Refactoring: code in filter.c used (void *) for "any old type" but this
4639 gives compiler warnings in some environments. I've now done it "properly",
4642 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
4643 (because of problems with the built-in one) was declared to return uschar *
4644 instead of char *, causing compiler failure.
4646 25. Fixed a file descriptor leak when processing alias/forward files.
4648 26. Fixed a minor format string issue in dbfn.c.
4650 27. Typo in exim.c: ("dmbnz" for "dbmnz").
4652 28. If a filter file refered to $h_xxx or $message_headers, and the headers
4653 contained RFC 2047 "words", Exim's memory could, under certain conditions,
4656 29. When a sender address is verified, it is cached, to save repeating the test
4657 when there is more than one recipient in a message. However, when the
4658 verification involves a callout, it is possible for different callout
4659 options to be set for different recipients. It is too complicated to keep
4660 track of this in the cache, so now Exim always runs a verification when a
4661 callout is required, relying on the callout cache for the optimization.
4662 The overhead is duplication of the address routing, but this should not be
4665 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
4666 to be verified with callout=postmaster, and the main callout worked but the
4667 postmaster check failed, the verification correctly failed. However, if a
4668 subsequent RCPT command asked for sender verification *without* the
4669 postmaster check, incorrect caching caused this verification also to fail,
4672 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
4673 it was not caching the DNS options (qualify_single, search_parents) that
4674 were used when the lookup failed. A subsequent lookup with different
4675 options therefore always gave the same answer, though there were cases
4676 where it should not have. (Example: a "domains = !$mx_any" option on a
4677 dnslookup router: the "domains" option is always processed without any
4678 widening, but the router might have qualify_single set.) Now Exim uses the
4679 cached value only when the same options are set.
4681 32. Added John Jetmore's "exipick" utility to the distribution.
4683 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
4684 than a timeout (e.g. a certificate is required, and is not provided), an
4685 Exim server now closes the connection immediately. Previously it waited for
4686 the client to close - but if the client is SSL, it seems that they each
4687 wait for each other, leading to a delay before one of them times out.
4689 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
4690 maintained 0.8.x compatibility because I don't think many are using it, and
4691 it is clearly obsolete.
4693 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
4696 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
4697 client certificate was expired. A simple patch fixes this, though I don't
4698 understand the full logic of why the verify callback is called multiple
4701 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
4702 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
4703 which causes problems with some clients (such as the Certicom SSL Plus
4704 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
4705 disables the coutermeasure allowing Eudora to connect."
4707 38. Exim was not checking that a write() to a log file succeeded. This could
4708 lead to Bad Things if a log got too big, in particular if it hit a file
4709 size limit. Exim now panics and dies if it cannot write to a log file, just
4710 as it does if it cannot open a log file.
4712 39. Modified OS/Makefile-Linux so that it now contains
4714 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
4716 The two -D definitions ensure that Exim is compiled with large file
4717 support, which makes it possible to handle log files that are bigger than
4720 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
4721 instance) a domain was checked against a named list that involved a lookup,
4722 causing $domain_data to be set, then another domain was checked against the
4723 same list, then the first domain was re-checked, the value of $domain_data
4724 after the final check could be wrong. In particular, if the second check
4725 failed, it could be set empty. This bug probably also applied to
4728 41. The strip_trailing_dot option was not being applied to the address given
4729 with the -f command-line option.
4731 42. The code for reading a message's header from the spool was incrementing
4732 $received_count, but never initializing it. This meant that the value was
4733 incorrect (doubled) while delivering a message in the same process in which
4734 it was received. In the most common configuration of Exim, this never
4735 happens - a fresh exec is done - but it can happen when
4736 deliver_drop_privilege is set.
4738 43. When Exim logs an SMTP synchronization error - client data sent too soon -
4739 it now includes up to 150 characters of the unexpected data in the log
4742 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
4743 and building data strings. The size of both of these buffers was 10 000
4744 bytes - far larger than anybody would *ever* want, thought I. Needless to
4745 say, somebody hit the limit. I have increased the maximum line length to
4746 20 000 and the maximum data length of concatenated lines to 100 000. I have
4747 also fixed two bugs, because there was no checking on these buffers. Tsk,
4748 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
4749 buffer is too small.
4751 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
4752 lsearch lookups. Now it does.
4754 46. When parsing a route_list item in a manualroute router, a fixed-length
4755 buffer was used for the list of hosts. I made this 1024 bytes long,
4756 thinking that nobody would ever have a list of hosts that long. Wrong.
4757 Somebody had a whole pile of complicated expansion conditions, and the
4758 string was silently truncated, leading to an expansion error. It turns out
4759 that it is easier to change to an unlimited length (owing to other changes
4760 that have happened since this code was originally written) than to build
4761 structure for giving a limitation error. The length of the item that
4762 expands into the list of hosts is now unlimited.
4764 47. The lsearch lookup could not handle data where the length of text line was
4765 more than 4095 characters. Such lines were truncated, leading to shortened
4766 data being returned. It should now handle lines of any length.
4768 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
4769 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
4772 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
4773 replacing strings such as BIN_DIRECTORY in a source file: the replacement
4774 no longer happens in comment lines. A list of replacements is now placed
4775 at the head of all of the source files, except those whose only change is
4776 to replace PERL_COMMAND in the very first #! line.
4778 50. Replaced the slow insertion sort in queue.c, for sorting the list of
4779 messages on the queue, with a bottom-up merge sort, using code contributed
4780 by Michael Haardt. This should make operations like -bp somewhat faster on
4781 large queues. It won't affect queue runners, except when queue_run_in_order
4784 51. Installed eximstats 1.31 in the distribution.
4786 52. Added support for SRV lookups to the dnslookup router.
4788 53. If an ACL referred to $message_body or $message_body_end, the value was not
4789 reset for any messages that followed in the same SMTP session.
4791 54. The store-handling optimization for building very long strings was not
4792 differentiating between the different store pools. I don't think this
4793 actually made any difference in practice, but I've tidied it.
4795 55. While running the routers to verify a sender address, $sender_address
4796 was still set to the sender address. This is wrong, because when routing to
4797 send a bounce to the sender, it would be empty. Therefore, I have changed
4798 it so that, while verifying a sender address, $sender_address is set to <>.
4799 (There is no change to what happens when verifying a recipient address.)
4801 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
4802 target A or AAAA records (if not already returned) without resetting the
4803 qualify_single or search_parents options of the DNS resolver. These are
4804 inappropriate in this case because the targets of MX and SRV records must
4805 be FQDNs. A broken DNS record could cause trouble if it happened to have a
4806 target that, when qualified, matched something in the local domain. These
4807 two options are now turned off when doing these lookups.
4809 57. It seems that at least some releases of Reiserfs (which does not have the
4810 concept of a fixed number of inodes) returns zero and not -1 for the
4811 number of available inodes. This interacted badly with check_spool_inodes,
4812 which assumed that -1 was the "no such thing" setting. What I have done is
4813 to check that the total number of inodes is greater than zero before doing
4814 the test of how many are available.
4816 58. When a "warn" ACL statement has a log_message modifier, the message is
4817 remembered, and not repeated. This is to avoid a lot of repetition when a
4818 message has many recipients that cause the same warning to be written.
4819 Howewer, Exim was preserving the list of already written lines for an
4820 entire SMTP session, which doesn't seem right. The memory is now reset if a
4821 new message is started.
4823 59. The "rewrite" debugging flag was not showing the result of rewriting in the
4824 debugging output unless log_rewrite was also set.
4826 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
4827 of (int)(handle) when we know that handle contains (void *)(-1).
4829 61. The Exim daemon panic-logs an error return when it closes the incoming
4830 connection. However "connection reset by peer" seems to be common, and
4831 isn't really an error worthy of noting specially, so that particular error
4834 62. When Exim is trying to find all the local interfaces, it used to panic and
4835 die if the ioctl to get the interface flags failed. However, it seems that
4836 on at least one OS (Solaris 9) it is possible to have an interface that is
4837 included in the list of interfaces, but for which you get a failure error
4838 for this call. This happens when the interface is not "plumbed" into a
4839 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
4840 failure of the "get flags" call assumes that the interface is down.
4842 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
4843 makes life easier for people who are doing arithmetic on fields extracted
4844 from dates, where you often get leading zeros that should not be
4845 interpreted as octal.
4847 64. Added qualify_domain to the redirect router, to override the global
4850 65. If a pathologically long header line contained very many addresses (the
4851 report of this problem mentioned 10 000) and each of them was rewritten,
4852 Exim could use up a very large amount of memory. (It kept on making new
4853 copies of the header line as it rewrote, and never released the old ones.)
4854 At the expense of a bit more processing, the header rewriting function has
4855 been changed so that it no longer eats memory in this way.
4857 66. The generation of the Received: header has been moved from the time that a
4858 message starts to be received, to the time that it finishes. The timestamp
4859 in the Received: header should now be very close to that of the <= log
4860 line. There are two side-effects of this change:
4862 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
4863 logged header lines no longer include the local Received: line, because
4864 it has not yet been created. The same applies to a copy of the message
4865 that is returned to a non-SMTP sender when a message is rejected.
4867 (b) When a filter file is tested using -bf, no additional Received: header
4868 is added to the test message. After some thought, I decided that this
4871 This change does not affect the value of $received_for. It is still set
4872 after address rewriting, but before local_scan() is called.
4874 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
4876 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
4877 gave an unhelpful panic error message, and a defer error. I have managed to
4878 change this behaviour so that it now rejects any supplied certificate,
4879 which seems right, as the list of acceptable certificates is empty.
4881 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
4882 gave an unhelpful defer error. I have not managed to make this reject any
4883 supplied certificates, but the error message it gives is "no certificate
4884 supplied", which is not helpful.
4886 70. exigrep's output now also includes lines that are not associated with any
4887 message, but which match the given pattern. Implemented by a patch from
4888 Martin Sluka, which also tidied up the Perl a bit.
4890 71. Recipient callout verification, like sender verification, was using <> in
4891 the MAIL FROM command. This isn't really the right thing, since the actual
4892 sender may affect whether the remote host accepts the recipient or not. I
4893 have changed it to use the actual sender in the callout; this means that
4894 the cache record is now keyed on a recipient/sender pair, not just the
4895 recipient address. There doesn't seem to be a real danger of callout loops,
4896 since a callout by the remote host to check the sender would use <>.
4897 [SEE ABOVE: changed after hitting problems.]
4899 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
4900 temporary errors. However, in the case of such a code being given after
4901 the end of a data transmission (i.e. after ".") Exim was failing to write
4902 a retry record for the message. (Yes, there was some broken host that was
4903 actually sending 8xx at this point.)
4905 73. An unknown lookup type in a host list could cause Exim to panic-die when
4906 the list was checked. (An example that provoked this was putting <; in the
4907 middle of a list instead of at the start.) If this happened during a DATA
4908 ACL check, a -D file could be left lying around. This kind of configuration
4909 error no longer causes Exim to die; instead it causes a defer errror. The
4910 incident is still logged to the main and panic logs.
4912 74. Buglet left over from Exim 3 conversion. The message "too many messages
4913 in one connection" was written to the rejectlog but not the mainlog, except
4914 when address rewriting (yes!) was being logged.
4916 75. Added write_rejectlog option.
4918 76. When a system filter was run not as root (that is, when system_filter_user
4919 was set), the values of the $n variables were not being returned to the
4920 main process; thus, they were not subsequently available in the $sn
4923 77. Added +return_path_on_delivery log selector.
4925 78. A connection timeout was being treated differently from recipients deferred
4926 when testing hosts_max_try with a message that was older than the host's
4927 retry timeout. (The host should not be counted, thus allowing all hosts to
4928 be tried at least once before bouncing.) This may have been the cause of an
4929 occasionally reported bug whereby a message would remain on the queue
4930 longer than the retry timeout, but would be bounced if a delivery was
4931 forced. I say "may" because I never totally pinned down the problem;
4932 setting up timeout/retry tests is difficult. See also the next item.
4934 79. The ultimate address timeout was not being applied to errors that involved
4935 a combination of host plus message (for example, a timeout on a MAIL
4936 command). When an address resolved to a number of possible hosts, and they
4937 were not all tried for each delivery (e.g. because of hosts_max_try), a
4938 message could remain on the queue longer than the retry timeout.
4940 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4943 81. Fixed an obscure SMTP outgoing bug which required at least the following
4944 conditions: (a) there was another message waiting for the same server;
4945 (b) the server returned 5xx to all RCPT commands in the first message so
4946 that the message was not completed; (c) the server dropped the connection
4947 or gave a negative response to the RSET that Exim sends to abort the
4948 transaction. The observed case was a dropped connection after DATA that had
4949 been sent in pipelining mode. That is, the server had advertised PIPELINING
4950 but was not implementing it correctly. The effect of the bug was incorrect
4951 behaviour, such as trying another host, and this could lead to a crash.
4957 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4958 and daemon.c were passed as pointers to ints; they should have been
4959 pointers to socklen_t variables (which are typically unsigned ints).
4961 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4964 3. Fixed a really odd bug that affected only the testing scheme; patching a
4965 certain fixed string in the binary changed the value of another string that
4966 happened to be identical to the end of the original first string.
4968 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4969 name", it returns that address as the IP address. On some operating
4970 systems (e.g. Solaris), it also passes back the IP address string as the
4971 "host name". However, on others (e.g. Linux), it passes back an empty
4972 string. Exim wasn't checking for this, and was changing the host name to an
4973 empty string, assuming it had been canonicized.
4975 5. Although rare, it is permitted to have more than one PTR record for a given
4976 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4977 all the names associated with an address, because they do in Solaris.
4978 However, it seems that they do not in Linux for data that comes from the
4979 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4980 I found this out when I moved to a new Linux workstation and tried to run
4981 the Exim test suite.
4983 To get round this problem I have changed the code so that it now does its
4984 own call to the DNS to look up PTR records when searching for a host name.
4985 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
4986 addresses that are only in /etc/hosts are still found.
4988 This behaviour is, however, controlled by an option called host_lookup_
4989 order, which defaults to "bydns:byaddr". If people want to use the other
4990 order, or indeed, just use one or the other means of lookup, they can
4991 specify it in this variable.
4993 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
4994 some operating systems, this comes back from gethostbyaddr() as an empty
4995 string, and this is what Exim used to test for. However, it seems that in
4996 other systems, "." is yielded. Exim now tests for this case too.
4998 7. The values of check_spool_space and check_log_space are now held internally
4999 as a number of kilobytes instead of an absolute number of bytes. If a
5000 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5001 kilobyte. This means that much larger values can be stored.
5003 8. Exim monitor: an attempt to get the action menu when not actually pointing
5004 at a message produces an empty menu entitled "No message selected". This
5005 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5006 no entries in it ("Shell widget menu has zero width and/or height"). So I
5007 have added a single, blank menu entry in this case.
5009 9. Added ${quote_local_part.
5011 10. MIME decoding is now applied to the contents of Subject: header lines when
5014 11. Now that a reference to $sender_host_address automatically causes a reverse
5015 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5016 host lookup before query-style lookups in lists that might use this
5017 variable. This has therefore been abolished, and the "net-" prefix is no
5018 longer necessary for query-style lookups.
5020 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5021 have been a typo for LFLAGS, so it has been changed.
5023 13. The install script calls Exim with "-C /dev/null" in order to find the
5024 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5025 to be output. Howeve, since Exim outputs its version number before the
5026 error, it didn't break the script. It just looked ugly. I fixed this by
5027 always allowing "-C /dev/null" if the caller is root.
5029 14. Ignore overlarge ACL variable number when reading spool file - insurance
5030 against a later release with more variables having written the file.
5032 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5033 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5034 this, as well as the form without the "IPv6" on the front.
5036 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5037 OS/Makefile-Darwin file.
5039 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5040 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5042 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5043 variable. However, some OS use ints for inodes. I've added cast to long int
5044 to get rid of the compiler warning.
5046 19. I had forgotten to lock out "/../" in configuration file names when
5047 ALT_CONFIG_PREFIX was set.
5049 20. Routers used for verification do not need to specify transports. However,
5050 if such a router generated a host list, and callout was configured, Exim
5051 crashed, because it could not find a port number from the (non-existent)
5052 transport. It now assumes port 25 in this circumstance.
5054 21. Added the -t option to exigrep.
5056 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5057 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5058 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5059 them is set, LOOKUP_LSEARCH is forced.
5061 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5063 24. Added sender and host information to the "rejected by local_scan()" log
5064 line; previously there was no indication of these.
5066 25. Added .include_if_exists.
5068 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5069 Linux. It turns out that not all file systems support this. Apparently some
5070 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5071 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5072 not supported on the file descriptor, is now ignored when Exim is trying to
5073 sync a directory. This applies only to Linux.
5075 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5077 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5078 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5081 29. Added support for maildirsize files from supplied patch (modified a bit).
5083 30. The use of :fail: followed by an empty string could lead Exim to respond to
5084 sender verification failures with (e.g.):
5086 550 Verification failed for <xxx>
5087 550 Sender verify failed
5089 where the first response line was missing the '-' that indicates it is not
5090 the final line of the response.
5092 31. The loop for finding the name of the user that called Exim had a hardwired
5093 limit of 10; it now uses the value of finduser_retries, which is used for
5094 all other user lookups.
5096 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5099 33. Exim was neglecting to zero errno before one call of strtol() when
5100 expanding a string and expecting an integer value. On some systems this
5101 resulted in spurious "integer overflow" errors. Also, it was casting the
5102 result into an int without checking.
5104 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5105 did not work. The code looks as if it has *never* worked, though it appears
5106 to have been documented since at least releast 1.62. I have made it work.
5108 35. The "timeout_DNS" error in retry rules, also documented since at least
5109 1.62, also never worked. As it isn't clear exactly what this means, and
5110 clearly it isn't a major issue, I have abolished the feature by treating it
5111 as "timeout", and writing a warning to the main and panic logs.
5113 36. The display of retry rules for -brt wasn't always showing the error code
5116 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5117 timeout_connect_A, timeout_connect_MX.
5119 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5120 to the empty sender.
5122 39. The daemon was not analysing the content of -oX till after it had closed
5123 stderr and disconnected from the controlling terminal. This meant that any
5124 syntax errors were only noted on the panic log, and the return code from
5125 the command was 0. By re-arranging the code a little, I've made the
5126 decoding happen first, so such errors now appear on stderr, and the return
5127 code is 1. However, the actual setting up of the sockets still happens in
5128 the disconnected process, so errors there are still only recorded on the
5131 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5132 connections (as happens on some IP stacks) was logged at start up time as
5133 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5134 it from "IPv6 and IPv4", which means that two separate sockets are being
5137 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5138 says whether AF_INET or AF_INET6 was passed as an argument.
5140 42. Exiwhat output was messed up when time zones were included in log
5143 43. Exiwhat now gives more information about the daemon's listening ports,
5144 and whether -tls-on-connect was used.
5146 44. The "port" option of the smtp transport is now expanded.
5148 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5149 silently ignored. Now an error message is written to the main and panic
5152 46. There's a new ACL modifier called "logwrite" which writes to a log file
5153 as soon as it is encountered.
5155 47. Added $local_user_uid and $local_user_gid at routing time.
5157 48. Exim crashed when trying to verify a sender address that was being
5160 49. Exim was recognizing only a space character after ".include". It now also
5161 recognizes a tab character.
5163 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5164 extracting the relevant information from the specification. The man page no
5165 longer contains scrambled data for the -d option, and I've added a section
5166 at the front about calling Exim under different names.
5168 51. Added "extra_headers" argument to the "mail" command in filter files.
5170 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5173 53. Installed eximstats 1.29.
5175 54. Added transport_filter_timeout as a generic transport option.
5177 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5178 Cc: header lines. This was required by RFC 822, but it not required by RFC
5181 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5182 incoming messages that did not have them. Now it does so only if the
5183 message originates locally, that is, if there is no associated remote host
5184 address. When Resent- header lines are present, this applies to the Resent-
5185 lines rather than the non-Resent- lines.
5187 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5188 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5190 58. Messages for configuration errors now include the name of the main
5191 configuration file - useful now that there may be more than one file in a
5192 list (.included file names were always shown).
5194 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5195 for those rare installations that do not start the daemon as root or run it
5196 setuid root. I've cut out the call to initgroups() if the daemon is not
5199 60. The Exim user and group can now be bound into the binary as text strings
5200 that are looked up at the start of Exim's processing.
5202 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5204 62. Added $mailstore_basename variable.
5206 63. Installed patch to sieve.c from Michael Haardt.
5208 64. When Exim failed to open the panic log after failing to open the main log,
5209 the original message it was trying to log was written to stderr and debug
5210 output, but if they were not available (the usual case in production), it
5211 was lost. Now it is written to syslog before the two lines that record the
5212 failures to open the logs.
5214 65. Users' Exim filters run in subprocesses under the user's uid. It is
5215 possible for a "deliver" command or an alias in a "personal" command to
5216 provoke an address rewrite. If logging of address rewriting is configured,
5217 this fails because the process is not running as root or exim. There may be
5218 a better way of dealing with this, but for the moment (because 4.30 needs
5219 to be released), I have disabled address rewrite logging when running a
5220 filter in a non-root, non-exim process.
5226 1. The buildconfig auxiliary program wasn't quoting the value set for
5227 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5228 not defined. This bug was masked in 4.22 by the effect that was fixed in
5231 2. Some messages that were rejected after a message id was allocated were
5232 shown as "incomplete" by exigrep. It no longer does this for messages that
5233 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5235 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5236 have allow_domain_literals set, the ID did not get logged in the <= line.
5237 Domain literals are now always recognized in Message-ID: header lines.
5239 4. The first argument for a ${extract expansion item is the key name or field
5240 number. Leading and trailing spaces in this item were not being ignored,
5241 causing some misleading effects.
5243 5. When deliver_drop_privilege was set, single queue runner processes started
5244 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5245 same command in the process it spins off) were not dropping privilege.
5247 6. When the daemon running as "exim" started a queue runner, it always
5248 re-executed Exim in the spun-off process. This is a waste of effort when
5249 deliver_drop_privilege is set. The new process now just calls the
5250 queue-runner function directly.
5256 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5259 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5260 ignored. Though the use of -oP was forcing the writing of a pid file, it
5261 was always written to the default place.
5263 3. If the message "no IP address found for host xxxx" is generated during
5264 incoming verification, it is now followed by identification of the incoming
5265 connection (so you can more easily find what provoked it).
5267 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5269 5. Added some features to "harden" Exim a bit more against certain attacks:
5271 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5272 be put in Local/Makefile. This is like the never_users runtime option,
5273 but it cannot be overridden. The default setting is "root".
5275 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5276 prefix string with which any file named in a -C command line option
5279 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5280 is retained for -C and -D only if the caller of Exim is root. Without
5281 it, the exim user may also use -C and -D and retain privilege.
5283 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5284 command line option is disabled.
5286 6. Macro names set by the -D option must start with an upper case letter, just
5287 like macro names defined in the configuration file.
5289 7. Added "dereference=" facility to LDAP.
5291 8. Two instances of the typo "uknown" in the source files are fixed.
5293 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5294 the Configure-Makefile script screwed up while processing it.
5296 10. Incorporated PCRE 4.4.
5298 11. The SMTP synchronization check was not operating right at the start of an
5299 SMTP session. For example, it could not catch a HELO sent before the client
5300 waited for the greeting. There is now a check for outstanding input at the
5301 point when the greeting is written. Because of the duplex, asynchronous
5302 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5303 way, but not yet received, when the check is performed.
5305 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5306 on TCP/IP sockets, because this apparently causes some broken clients to
5309 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5310 unchanged) from the Cygwin maintainer.
5312 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5313 was supported instead of testing for mbx. Effectively a typo.
5315 15. The spa authenticator server code was not checking that the input it
5316 received was valid base64.
5318 16. The debug output line for the "set" modifier in ACLs was not showing the
5319 name of the variable that was being set.
5321 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5323 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5324 lookup on its own. Something else was needed to trigger the lookup. For
5325 example, a match in host_lookup or the need for a host name in a host list.
5326 Now, if $sender_host_name is referenced and the host name has not yet been
5327 looked up, a lookup is performed. If the lookup fails, the variable remains
5328 empty, and $host_lookup_failed is set to "1".
5330 19. Added "eqi" as a case-independent comparison operator.
5332 20. The saslauthd authentication condition could segfault if neither service
5333 nor realm was specified.
5335 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5336 error message that was logged was misleading, and incoming SMTP
5337 connections were dropped. The message is now more accurate, and temporary
5338 errors are given to SMTP connections.
5340 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5341 (except RSET) with a 421 error, until QUIT is received. However, it was
5342 failing to send a response to QUIT.
5344 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5345 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5346 ACL is now run after the helo_try_verify_hosts code.
5348 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5349 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5350 case-independent, so other case variants are also recognized). Apparently
5351 some systems use these upper case variants.
5353 25. If more than two messages were waiting for the same host, and a transport
5354 filter was specified for the transport, Exim sent two messages over the
5355 same TCP/IP connection, and then failed with "socket operation on non-
5356 socket" when it tried to send the third.
5358 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5360 27. The extern definition of crypt16() in expand.c was not being excluded when
5361 the OS had its own crypt16() function.
5363 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5364 as a preferred synonym for return_size_limit, both as an option and as an
5367 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5369 30. Changed the default configuration ACL to relax the local part checking rule
5370 for addresses that are not in any local domains. For these addresses,
5371 slashes and pipe symbols are allowed within local parts, but the sequence
5372 /../ is explicitly forbidden.
5374 31. SPA server authentication was not clearing the challenge buffer before
5377 32. log_message in a "warn" ACL statement was writing to the reject log as
5378 well as to the main log, which contradicts the documentation and doesn't
5379 seem right (because no rejection is happening). So I have stopped it.
5381 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5382 However, I am unable to do any testing of this.
5384 34. Fixed an infelicity in the appendfile transport. When checking directories
5385 for a mailbox, to see if any needed to be created, it was accidentally
5386 using path names with one or more superfluous leading slashes; tracing
5387 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5389 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5390 contents are added to the log line that is written for every discarded
5391 recipient. (Previously a log_message setting was ignored.)
5393 36. The ${quote: operator now quotes the string if it is empty.
5395 37. The install script runs exim in order to find its version number. If for
5396 some reason other than non-existence or emptiness, which it checks, it
5397 could not run './exim', it was installing it with an empty version number,
5398 i.e. as "exim-". This error state is now caught, and the installation is
5401 38. An argument was missing from the function that creates an error message
5402 when Exim fails to connect to the socket for saslauthd authentication.
5403 This could cause Exim to crash, or give a corrupted message.
5405 39. Added isip, isip4, and isip6 to ${if conditions.
5407 40. The ACL variables $acl_xx are now saved with the message, and can be
5408 accessed later in routers, transports, and filters.
5410 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
5411 strings in the file are not string-expanded.
5413 42. If a MAIL command specified a SIZE value that was too large to fit into an
5414 int variable, the check against message_size_limit failed. Such values are
5415 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
5416 day this will have to be increased, but I don't think I want to be around
5417 when emails are that large.
5424 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
5425 iconv() is not standard in FreeBSD.
5427 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
5428 IPv6 enabled. The observed symptom was a segmentation fault on return from
5429 the function os_common_find_running_interfaces() in src/os.c.
5431 3. In the check_special_case() function in daemon.c I had used "errno" as an
5432 argument name, which causes warnings on some systems. This was basically a
5433 typo, since it was named "eno" in the comments!
5435 4. The code that waits for the clock to tick (at a resolution of some fraction
5436 of a second) so as to ensure message-id uniqueness was always waiting for
5437 at least one whole tick, when it could have waited for less. [This is
5438 almost certainly not relevant at current processor speeds, where it is
5439 unlikely to ever wait at all. But we try to future-proof.]
5441 5. The function that sleeps for a time interval that includes fractions of a
5442 second contained a race. It did not block SIGALRM between setting the
5443 timer, and suspending (a couple of lines later). If the interval was short
5444 and the sigsuspend() was delayed until after it had expired, the suspension
5445 never ended. On busy systems this could lead to processes getting stuck for
5448 6. Some uncommon configurations may cause a lookup to happen in a queue runner
5449 process, before it forks any delivery processes. The open lookup caching
5450 mechanism meant that the open file or database connection was passed into
5451 the delivery process. The problem was that delivery processes always tidy
5452 up cached lookup data. This could cause a problem for the next delivery
5453 process started by the queue runner, because the external queue runner
5454 process does not know about the closure. So the next delivery process
5455 still has data in the lookup cache. In the case of a file lookup, there was
5456 no problem because closing a file descriptor in a subprocess doesn't affect
5457 the parent. However, if the lookup was caching a connection to a database,
5458 the connection was closed, and the second delivery process was likely to
5459 see errors such as "PGSQL: query failed: server closed the connection
5460 unexpectedly". The problem has been fixed by closing all cached lookups
5461 in a queue runner before running a delivery process.
5463 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
5464 seem to have the "const" qualifier which it has on other OS. I've
5467 8. Change 4.21/2 was too strict. It is only if there are two authenticators
5468 *of the same type* (client or server) with the same public name that an
5469 error should be diagnosed.
5471 9. When Exim looked up a host name for an IP address, but failed to find the
5472 original IP address when looking up the host name (a safety check), it
5473 output the message "<ip address> does not match any IP for NULL", which was
5474 confusing, to say the least. The bug was that the host name should have
5475 appeared instead of "NULL".
5477 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
5478 user that is built into the binary, and the -C or -D options is used, root
5479 privilege is dropped before the configuration file is read. In addition,
5480 logging is switched to stderr instead of the normal log files. If the
5481 configuration then re-defines the Exim user, the unprivileged environment
5482 is probably not what is expected, so Exim logs a panic warning message (but
5485 However, if deliver_drop_privilege is set, the unprivileged state may well
5486 be exactly what is intended, so the warning has been cut out in that case,
5487 and Exim is allowed to try to write to its normal log files.
5493 1. smtp_return_error_details was not giving details for temporary sender
5494 or receiver verification errors.
5496 2. Diagnose a configuration error if two authenticators have the same public
5499 3. Exim used not to create the message log file for a message until the first
5500 delivery attempt. This could be confusing when incoming messages were held
5501 for policy or load reasons. The message log file is now created at the time
5502 the message is received, and an initial "Received" line is written to it.
5504 4. The automatically generated man page for command line options had a minor
5505 bug that caused no ill effects; however, a more serious problem was that
5506 the procedure for building the man page automatically didn't always
5507 operate. Consequently, release 4.20 contains an out-of-date version. This
5508 shouldn't happen again.
5510 5. When building Exim with embedded Perl support, the script that builds the
5511 Makefile was calling 'perl' to find its compile-time parameters, ignoring
5512 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
5514 6. The freeze_tell option was not being used for messages that were frozen on
5515 arrival, either by an ACL or by local_scan().
5517 7. Added the smtp_incomplete_transaction log selector.
5519 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
5520 was accepting AUTH without a new EHLO.
5522 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
5523 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
5526 10. Logging of TCP/IP connections (when configured) now happens in the main
5527 daemon process instead of the child process, so that the TCP/IP connection
5528 count is more accurate (but it can never be perfect).
5530 11. The use of "drop" in a nested ACL was not being handled correctly in the
5531 outer ACL. Now, if condition failure induced by the nested "drop" causes
5532 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
5533 or "require"), the connection is dropped.
5535 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
5536 that yield "discard" can now be used with an "accept" or a "discard" verb,
5537 but an error is generated for any others (because I can't see a useful way
5538 to define what should happen).
5540 13. When an ACL is read dynamically from a file (or anywhere else), the lines
5541 are now processed in the same way as lines in the Exim configuration file.
5542 In particular, continuation lines are supported.
5544 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
5546 15. Added -ti meaning -t -i.
5548 16. Check for letters, digits, hyphens, and dots in the names of dnslist
5549 domains, and warn by logging if others are found.
5551 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
5552 returned from GIFCONF when Exim is trying to find the list of interfaces on
5553 a host. The code in os.c has been modified to copy each ifreq to an aligned
5554 structure in all cases.
5556 Also, in some cases, the returned ifreq's were being copied to a 'struct
5557 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
5558 means the last couple of bytes of an IPv6 address could be chopped if the
5559 ifreq contained only a normal sockaddr (14 bytes storage).
5561 18. Named domain lists were not supported in the hosts_treat_as_local option.
5562 An entry such as +xxxx was not recognized, and was treated as a literal
5565 19. Ensure that header lines added by a DATA ACL are included in the reject log
5566 if the ACL subsequently rejects the message.
5568 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
5569 MD5 (which is deprecated).
5571 21. When testing a filter file using -bf, Exim was writing a message when it
5572 took the sender from a "From " line in the message, but it was not doing so
5573 when it took $return_path from a Return-Path: header line. It now does.
5575 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
5576 with a valid header line field name (a series of printing characters
5577 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
5579 23. Changed "disc" in the source to "disk" to conform to the documentation and
5580 the book and for uniformity.
5582 24. Ignore Sendmail's -Ooption=value command line item.
5584 25. When execve() failed while trying to run a command in a pipe transport,
5585 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
5586 could be confused with a return value of 69 from the command itself. This
5587 has been changed to 127, the value the shell returns if it is asked to run
5588 a non-existent command. The wording for the related log line suggests a
5589 non-existent command as the problem.
5591 26. If received_header_text expands to an empty string, do not add a Received:
5592 header line to the message. (Well, it adds a token one on the spool, but
5593 marks it "old" so that it doesn't get used or transmitted.)
5595 27. Installed eximstats 1.28 (addition of -nt option).
5597 28. There was no check for failure on the call to getsockname() in the daemon
5598 code. This can fail if there is a shortage of resources on the system, with
5599 ENOMEM, for example. A temporary error is now given on failure.
5601 29. Contrary to the C standard, it seems that in some environments, the
5602 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
5603 program. Exim now does this explicitly; it affects the formatting of
5604 timestamps using strftime().
5606 30. If exiqsumm was given junk data, it threw up some uninitialized variable
5607 complaints. I've now initialized all the variables, to avoid this.
5609 32. Header lines added by a system filter were not being "seen" during
5610 transport-time rewrites.
5612 33. The info_callback() function passed to OpenSSL is set up with type void
5613 (*)(SSL *, int, int), as described somewhere. However, when calling the
5614 function (actually a macro) that sets it up, the type void(*)() is
5615 expected. I've put in a cast to prevent warnings from picky compilers.
5617 34. If a DNS black list lookup found a CNAME record, but there were no A
5618 records associated with the domain it pointed at, Exim crashed.
5620 35. If a DNS black list lookup returned more than one A record, Exim ignored
5621 all but the first. It now scans all returned addresses if a particular IP
5622 value is being sought. In this situation, the contents of the
5623 $dnslist_value variable are a list of all the addresses, separated by a
5626 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
5627 to accept a host name and all its aliases if the forward lookup for any of
5628 them yielded the IP address of the incoming connection. Now it accepts only
5629 those names whose forward lookup yields the correct IP address. Any other
5630 names are discarded. This closes a loophole whereby a rogue DNS
5631 administrator could create reverse DNS records to break through a
5632 wildcarded host restriction in an ACL.
5634 37. If a user filter or a system filter that ran in a subprocess used any of
5635 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
5636 the wrong values were passed to the pipe command ($thisaddress had the
5637 value of $0, $0 had the value of $1, etc). This bug was introduced by
5638 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
5640 38. Improved the line breaking for long SMTP error messages from ACLs.
5641 Previously, if there was no break point between 40 and 75 characters, Exim
5642 left the rest of the message alone. Two changes have been made: (a) I've
5643 reduced the minimum length to 35 characters; (b) if it can't find a break
5644 point between 35 and 75 characters, it looks ahead and uses the first one
5645 that it finds. This may give the occasional overlong line, but at least the
5646 remaining text gets split now.
5648 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
5649 file descriptors might be low, and that setting 1000 would always raise it.
5650 It turns out that in some environments, the limit is already over 1000 and
5651 that lowering it causes trouble. So now Exim takes care not to decrease it.
5653 40. When delivering a message, the value of $return_path is set to $sender_
5654 address at the start of routing (routers may change the value). By an
5655 oversight, this default was not being set up when an address was tested by
5656 -bt or -bv, which affected the outcome if any router or filter referred to
5659 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
5660 log only when "message" or "log_message" are set. However, if one of the
5661 conditions was an address verification, or a call to a nested ACL, the
5662 messages generated by the underlying test were being passed through. This
5663 no longer happens. The underlying message is available in $acl_verify_
5664 message for both "message" and "log_message" expansions, so it can be
5665 passed through if needed.
5667 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
5668 new expansion $bh_ to give the encoded byte string without charset
5669 translation. Translation happens only if iconv() is available; HAVE_ICONV
5670 indicates this at build time. HEADERS_CHARSET gives the charset to
5671 translate to; headers_charset can change it in the configuration, and
5672 "headers charset" can change it in an individual filter file.
5674 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
5675 that creates RFC 2047 encoded "words" labels them as that charset instead
5676 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
5677 expansion operator; (ii) when Exim creates a From: line for a local
5678 message; (iii) when a header line is rewritten to include a "phrase" part.
5680 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
5681 buggy, causing it to skip the first lines of messages whose message ID
5682 ended in 'D'. This would not have bitten before Exim release 4.14, because
5683 message IDs were unlikely to end in 'D' before then. The effect was to have
5684 incorrect size information for certain domains.
5686 45. #include "config.h" was missing at the start of the crypt16.c module. This
5687 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
5690 46. If there was a timeout during a "random" callout check, Exim treated it as
5691 a failure of the random address, and carried on sending RSET and the real
5692 address. If the delay was just some slowness somewhere, the response to the
5693 original RCPT would be taken as a response to RSET and so on, causing
5694 mayhem of various kinds.
5696 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
5697 when I implemented it. It didn't allow for the fact that some option values
5698 may legitimatetly be negative (e.g. size_addition), and it didn't even do
5699 the right test for positive values.
5701 48. Domain names in DNS records are case-independent. Exim always looks them up
5702 in lower case. Some resolvers return domain names in exactly the case they
5703 appear in the zone file, that is, they may contain uppercase letters. Not
5704 all resolvers do this - some return always lower case. Exim was treating a
5705 change of case by a resolver as a change of domain, similar to a widening
5706 of a domain abbreviation. This triggered its re-routing code and so it was
5707 trying to route what was effectively the same domain again. This normally
5708 caused routing to fail (because the router wouldn't handle the domain
5709 twice). Now Exim checks for this case specially, and just changes the
5710 casing of the domain that it ultimately uses when it transmits the message
5713 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
5716 50. If a filter generated a file delivery with a non-absolute name (possible if
5717 no home directory exists for the router), the forbid_file option was not
5720 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
5721 the existing equality matching.
5723 52. Exim was using ints instead of ino_t variables in some places where it was
5724 dealing with inode numbers.
5726 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
5727 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
5728 called TMPDIR, and if it finds it is different, it changes its value.
5730 54. The smtp_printf() function is now made available to local_scan() so
5731 additional output lines can be written before returning. There is also an
5732 smtp_fflush() function to enable the detection of a dropped connection.
5733 The variables smtp_input and smtp_batched_input are exported to
5736 55. Changed the default runtime configuration: the message "Unknown user"
5737 has been removed from the ACL, and instead placed on the localuser router,
5738 using the cannot_route_message feature. This means that any verification
5739 failures that generate their own messages won't get overridden. Similarly,
5740 the "Unrouteable address" message that was in the ACL for unverifiable
5741 relay addresses has also been removed.
5743 56. Added hosts_avoid_esmtp to the smtp transport.
5745 57. The exicyclog script was not checking for the esoteric option
5746 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
5747 will work only if exicyclog is run under the appropriate euid.
5749 58. Following a discussion on the list, the rules by which Exim recognises line
5750 endings on incoming messages have been changed. The -dropcr and drop_cr
5751 options are now no-ops, retained only for backwards compatibility. The
5752 following line terminators are recognized: LF CRLF CR. However, special
5753 processing applies to CR:
5755 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
5756 nor a local message in the state where . is a terminator.
5758 (ii) If a bare CR is encountered in a header line, an extra space is added
5759 after the line terminator so as not to end the header. The reasoning
5760 behind this is that bare CRs in header lines are most likely either
5761 to be mistakes, or people trying to play silly games.
5763 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
5764 was being incorrectly given as 18 bytes larger than it should have been.
5765 This is a VOB (very old bug).
5767 60. This may never have affected anything current, but just in case it has:
5768 When the local host is found other than at the start of a list of hosts,
5769 the local host, those with the same MX, and any that follow, are discarded.
5770 When the list in question was part of a longer list of hosts, the following
5771 hosts (not currently being processed) were also being discarded. This no
5772 longer happens. I'm not sure if this situation could ever has previously
5775 61. Added the "/MX" feature to lists of hosts in the manualroute and query
5778 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
5779 header. This is something that is recommended in a new Internet Draft, and
5780 is something that is documented as being done by Sendmail. There are two
5781 possible values. For messages generated by the autoreply transport, Exim
5784 Auto-Submitted: auto-replied
5786 whereas for all other generated messages (e.g. bounces) it adds
5788 Auto-Submitted: auto-generated
5790 63. The "personal" condition in filters now includes a test for the
5791 Auto-Submitted: header. If it contains the string "auto-" the message it
5792 not considered personal.
5794 64. Added rcpt_include_affixes as a generic transport option.
5796 65. Added queue_only_override (default true).
5798 66. Added the syslog_duplication option.
5800 67. If what should have been the first header line of a message consisted of
5801 a space followed by a colon, Exim was mis-interpreting it as a header line.
5802 It isn't of course - it is syntactically invalid and should therefore be
5803 treated as the start of the message body. The misbehaviour could have
5804 caused a number of strange effects, including loss of data in subsequent
5805 header lines, and spool format errors.
5807 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
5808 client host had authenticated. This control can now be exercised by an ACL
5809 for more flexibility.
5811 69. By default, callouts do not happen when testing with -bh. There is now a
5812 variant, -bhc, which does actually run the callout code, including
5813 consulting and updating the callout cache.
5815 70. Added support for saslauthd authentication, courtesy of Alexander
5818 71. If statvfs() failed on the spool or log directories while checking their
5819 size for availability, Exim confusingly gave the error "space shortage".
5820 Furthermore, in debugging mode it crashed with a floating point exception.
5821 These checks are done if check_{spool,log}_{space,inodes} are set, and when
5822 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
5823 serious problem, Exim now writes to the main and panic logs when this
5824 happens, with details of the failure. It then refuses to accept the
5825 incoming message, giving the message "spool directory problem" or "log
5826 directory problem" with a 421 code for SMTP messages.
5828 72. When Exim is about to re-exec itself, it ensures that the file descriptors
5829 0, 1, and 2 exist, because some OS complain for execs without them (see
5830 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
5831 descriptors. However, the code omitted to check that the open succeeded,
5832 causing mysterious errors if for some reason the permissions on /dev/null
5833 got screwed. Now Exim writes a message to the main and panic logs, and
5834 bombs out if it can't open /dev/null.
5836 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
5837 interact so that it is all more flexible. It is supposed to remain
5838 backwards compatible. Also added extra_local_interfaces.
5840 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
5841 to bomb out with an assertion failure - to the client this appears as a
5842 connection drop. This problem occurs in the part of the code that was taken
5843 from the Samba project. Fortunately, the assertion is in a very simple
5844 function, so I have fixed this by reproducing the function inline in the
5845 one place where it is called, and arranging for authentication to fail
5846 instead of killing the process with assert().
5848 75. The SPA client code was not working when the server requested OEM rather
5849 than Unicode encoding.
5851 76. Added code to make require_files with a specific uid setting more usable in
5852 the case where statting the file as root fails - usually a non-root-mounted
5853 NFS file system. When this happens and the failure is EACCES, Exim now
5854 forks a subprocess and does the per-uid checking as the relevant uid.
5856 77. Added process_log_path.
5858 78. If log_file_path was not explicitly set, a setting of check_log_space or
5859 check_log_inodes was ignored.
5861 79. If a space check for the spool or log partitions fails, the incident is now
5862 logged. Of course, in the latter case the data may get lost...
5864 80. Added the %p formatting code to string_format() so that it can be used to
5865 print addresses in debug_print(). Adjusted all the address printing in the
5866 debugging in store.c to use %p rather than %d.
5868 81. There was a concern that a line of code in smtp_in.c could overflow a
5869 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
5870 initially expressed, the concern was not well-founded, because trailing
5871 spaces are removed early. However, if the trailing spaces were followed by
5872 a NULL, they did not get removed, so the overflow was possible. Two fixes
5875 (a) I re-wrote the offending code in a cleaner fashion.
5876 (b) If an incoming SMTP command contains a NULL character, it is rejected
5879 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
5880 runs initgroups(), so that if the Exim user is in any additional groups,
5881 they will be used during message reception.
5887 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob