1 Change log file for Exim from version 4.21
2 -------------------------------------------
7 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
9 PP/02 Make -n do something, by making it not do something.
10 When combined with -bP, the name of an option is not output.
12 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
15 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
16 $sender_host_name and config options to manage this, and basic check
19 PP/05 DSCP support for outbound connections and control modifier for inbound.
21 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
22 (Only plugin which currently uses this is kerberos4, which nobody should
23 be using, but we should make it available and other future plugins might
24 conceivably use it, even though it would break NAT; stuff *should* be
25 using channel bindings instead).
27 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
28 name; added for Sendmail compatibility; requires admin caller.
29 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
30 just ignore it); requires trusted caller.
31 Also parse but ignore: -Ac -Am -X<logfile>
34 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
36 TL/02 Add +smtp_confirmation as a default logging option.
38 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
39 Patch by Magnus Holmgren from 2007-02-20.
41 TL/04 Bugzilla 1281 - Spec typo.
42 Bugzilla 1283 - Spec typo.
43 Bugzilla 1290 - Spec grammar fixes.
45 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
47 TL/06 Add Experimental DMARC support using libopendmarc libraries.
49 TL/07 Fix an out of order global option causing a segfault. Reported to dev
50 mailing list by by Dmitry Isaikin.
52 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
54 JH/02 Support "G" suffix to numbers in ${if comparisons.
56 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
58 NM/01 Bugzilla 1197 - Spec typo
59 Bugzilla 1196 - Spec examples corrections
61 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
63 PP/09 Add gnutls_enable_pkcs11 option.
65 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
66 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
68 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
69 "acl {{name}{arg}...}", and optional args on acl condition
72 JH/05 Permit multiple router/transport headers_add/remove lines.
74 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
76 JH/07 Avoid using a waiting database for a single-message-only transport.
77 Performance patch from Paul Fisher. Bugzilla 1262.
79 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
82 JH/09 Add $headers_added variable, with content from use of ACL modifier
83 add_header (but not yet added to the message). Bugzilla 199.
85 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
86 Pulled from Bugzilla 817 by Wolfgang Breyha.
88 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
90 (nb: this is the same fix as in Exim 4.80.1)
92 JH/11 Add A= logging on delivery lines, and a client_set_id option on
95 JH/12 Add optional authenticated_sender logging to A= and a log_selector
98 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
100 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
101 advertise SMTP AUTH mechanism to us, instead of a generic
102 protocol violation error. Also, make Exim more robust to bad
103 data from the Dovecot auth socket.
105 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
107 When a queue runner is handling a message, Exim first routes the
108 recipient addresses, during which it prunes them based on the retry
109 hints database. After that it attempts to deliver the message to
110 any remaining recipients. It then updates the hints database using
113 So if a recipient address works intermittently, it can get repeatedly
114 deferred at routing time. The retry hints record remains fresh so the
115 address never reaches the final cutoff time.
117 This is a fairly common occurrence when a user is bumping up against
118 their storage quota. Exim had some logic in its local delivery code
119 to deal with this. However it did not apply to per-recipient defers
120 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
122 This change adds a proper retry rule check during routing so that the
123 final cutoff time is checked against the message's age. We only do
124 this check if there is an address retry record and there is not a
125 domain retry record; this implies that previous attempts to handle
126 the address had the retry_use_local_parts option turned on. We use
127 this as an approximation for the destination being like a local
128 delivery, as in LMTP.
130 I suspect this new check makes the old local delivery cutoff check
131 redundant, but I have not verified this so I left the code in place.
133 TF/02 Correct gecos expansion when From: is a prefix of the username.
135 Test 0254 submits a message to Exim with the header
139 When I ran the test suite under the user fanf2, Exim expanded
140 the header to contain my full name, whereas it should have added
141 a Resent-Sender: header. It erroneously treats any prefix of the
142 username as equal to the username.
144 This change corrects that bug.
146 GF/01 DCC debug and logging tidyup
147 Error conditions log to paniclog rather than rejectlog.
148 Debug lines prefixed by "DCC: " to remove any ambiguity.
150 TF/03 Avoid unnecessary rebuilds of lookup-related code.
152 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
153 Bug spotted by Jeremy Harris; was flawed since initial commit.
154 Would have resulted in OCSP responses post-SNI triggering an Exim
155 NULL dereference and crash.
157 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
159 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
160 Bug detection, analysis and fix by Samuel Thibault.
161 Bugzilla 1331, Debian bug #698092.
163 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
165 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
166 Server implementation by Todd Lyons, client by JH.
167 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
168 config variable "prdr_enable" controls whether the server
169 advertises the facility. If the client requests PRDR a new
170 acl_data_smtp_prdr ACL is called once for each recipient, after
171 the body content is received and before the acl_smtp_data ACL.
172 The client is controlled by bolth of: a hosts_try_prdr option
173 on the smtp transport, and the server advertisement.
174 Default client logging of deliveries and rejections involving
175 PRDR are flagged with the string "PRDR".
177 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
178 fclose(). Diagnosis by Todd Lyons.
180 PP/17 Update configure.default to handle IPv6 localhost better.
181 Patch by Alain Williams (plus minor tweaks).
184 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
185 This is now consistent with GnuTLS, and is now documented: the
186 previous undocumented portable approach to treating the option as
187 unset was to force an expansion failure. That still works, and
188 an empty string is now equivalent.
190 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
191 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
192 not performing validation itself.
194 PP/20 Added force_command boolean option to pipe transport.
195 Patch from Nick Koston, of cPanel Inc.
197 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
200 TF/04 Added udpsend ACL modifer and hexquote expansion operator
202 PP/21 Fix eximon continuous updating with timestamped log-files.
203 Broken in a format-string cleanup in 4.80, missed when I repaired the
204 other false fix of the same issue.
205 Report and fix from Heiko Schlichting.
208 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
209 Report from Prashanth Katuri.
211 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
212 It's SecureTransport, so affects any MacOS clients which use the
213 system-integrated TLS libraries, including email clients.
215 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
216 using a MIME ACL for non-SMTP local injection.
217 Report and assistance in diagnosis by Warren Baker.
219 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
221 JH/16 Fix comparisons for 64b. Bugzilla 1385.
223 TL/09 Add expansion variable $authenticated_fail_id to keep track of
224 last id that failed so it may be referenced in subsequent ACL's.
226 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
229 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
230 ldap library initialization, allowing self-signed CA's to be
231 used. Also properly sets require_cert option later in code by
232 using NULL (global ldap config) instead of ldap handle (per
233 session). Bug diagnosis and testing by alxgomz.
235 TL/12 Enhanced documentation in the ratelimit.pl script provided in
236 the src/util/ subdirectory.
238 TL/13 Bug 1301 - Imported transport SQL logging patch from Axel Rau
239 renamed to Transport Post Delivery Action by Jeremy Harris, as
242 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
243 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
244 redis_servers = needs to be configured which will be used by the redis
245 lookup. Patch from Warren Baker, of The Packet Hub.
247 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
249 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
250 hostname or reverse DNS when processing a host list. Used suggestions
251 from multiple comments on this bug.
257 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
259 This, or similar/improved, will also be change PP/11 of 4.82.
265 PP/01 Handle short writes when writing local log-files.
266 In practice, only affects FreeBSD (8 onwards).
267 Bugzilla 1053, with thanks to Dmitry Isaikin.
269 NM/01 Bugzilla 949 - Documentation tweak
271 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
274 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
276 PP/02 Implemented gsasl authenticator.
278 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
280 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
281 `pkg-config foo` for cflags/libs.
283 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
284 with rest of GSASL and with heimdal_gssapi.
286 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
287 `pkg-config foo` for cflags/libs for the TLS implementation.
289 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
290 properties get this fed in as external SSF. A number of robustness
291 and debugging improvements to the cyrus_sasl authenticator.
293 PP/08 cyrus_sasl server now expands the server_realm option.
295 PP/09 Bugzilla 1214 - Log authentication information in reject log.
296 Patch by Jeremy Harris.
298 PP/10 Added dbmjz lookup type.
300 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
302 PP/12 MAIL args handles TAB as well as SP, for better interop with
303 non-compliant senders.
304 Analysis and variant patch by Todd Lyons.
306 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
307 Bug report from Lars Müller <lars@samba.org> (via SUSE),
308 Patch from Dirk Mueller <dmueller@suse.com>
310 PP/13 tls_peerdn now print-escaped for spool files.
311 Observed some $tls_peerdn in wild which contained \n, which resulted
312 in spool file corruption.
314 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
315 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
316 or write after TLS renegotiation, which otherwise led to messages
319 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
320 as a tracking header (ie: a signed header comes before the signature).
321 Patch from Wolfgang Breyha.
323 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
324 comma-sep list; embedded commas doubled.
326 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
328 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
330 Report and patch from Dmitry Banschikov.
332 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
333 Removed SSL_clear() after SSL_new() which led to protocol negotiation
334 failures. We appear to now support TLS1.1+ with Exim.
336 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
337 lets Exim select keys and certificates based upon TLS SNI from client.
338 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
339 before an outbound SMTP session. New log_selector, +tls_sni.
341 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
342 NULL dereference. Report and patch from Alun Jones.
344 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
345 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
346 Not seeing resolver debug output on NetBSD, but suspect this is a
347 resolver implementation change.
349 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
350 Left warnings. Added "eximon gdb" invocation mode.
352 PP/21 Defaulting "accept_8bitmime" to true, not false.
354 PP/22 Added -bw for inetd wait mode support.
356 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
357 locate the relevant includes and libraries. Made this the default.
359 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
360 Bugzilla 1246, report and most of solution from Tomasz Kusy.
362 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
363 This may cause build issues on older platforms.
365 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
366 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
367 gnutls_require_mac & gnutls_require_protocols (no longer supported).
368 Added SNI support via GnuTLS too.
369 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
371 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
373 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
374 Applied second patch from Janne, implementing suggestion to default
375 multiple-strings-in-record handling to match SPF spec.
377 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
379 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
380 read-only, out of scope).
381 Patch from Wolfgang Breyha, report from Stuart Northfield.
383 PP/29 Fix three issues highlighted by clang analyser static analysis.
384 Only crash-plausible issue would require the Cambridge-specific
385 iplookup router and a misconfiguration.
386 Report from Marcin Mirosław.
388 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
390 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
391 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
392 As part of this, removing so much warning spew let me fix some minor
393 real issues in debug logging.
395 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
396 assignment on my part. Fixed.
398 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
399 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
400 Janne Snabb (who went above and beyond: thank you).
402 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
403 string otherwise requires a connection and a bunch more work and it's
404 relatively easy to get wrong. Should also expose TLS library linkage
407 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
408 64-bit ${eval} (JH/03).
410 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
411 GNU libc to support some of the 64-bit stuff, should not lead to
412 conflicts. Defined before os.h is pulled in, so if a given platform
413 needs to override this, it can.
415 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
416 protection layer was required, which is not implemented.
417 Bugzilla 1254, patch from Wolfgang Breyha.
419 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
420 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
421 tls_dhparam take prime identifiers. Also unbreak combination of
422 OpenSSL+DH_params+TLSSNI.
424 PP/39 Disable SSLv2 by default in OpenSSL support.
430 PP/01 Solaris build fix for Oracle's LDAP libraries.
431 Bugzilla 1109, patch from Stephen Usher.
433 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
435 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
438 TF/02 Fix a couple more cases where we did not log the error message
439 when unlink() failed. See also change 4.74-TF/03.
441 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
442 lock up or crash if it happened to be inside a call to libc when it
443 got a SIGUSR1 from exiwhat.
445 The SIGUSR1 handler appends the current process status to the process
446 log which is later printed by exiwhat. It used to use the general
447 purpose logging code to do this, but several functions it calls are
448 not safe for signals.
450 The new output code in the SIGUSR1 handler is specific to the process
451 log, and simple enough that it's easy to inspect for signal safety.
452 Removing some special cases also simplifies the general logging code.
453 Removing the spurious timestamps from the process log simplifies
456 TF/04 Improved ratelimit ACL condition.
458 The /noupdate option has been deprecated in favour of /readonly which
459 has clearer semantics. The /leaky, /strict, and /readonly update modes
460 are mutually exclusive. The update mode is no longer included in the
461 database key; it just determines when the database is updated. (This
462 means that when you upgrde Exim will forget old rate measurements.)
464 Exim now checks that the per_* options are used with an update mode that
465 makes sense for the current ACL. For example, when Exim is processing a
466 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
467 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
468 must specify per_mail/readonly. If you omit the update mode it defaults to
469 /leaky where that makes sense (as before) or /readonly where required.
471 The /noupdate option is now undocumented but still supported for
472 backwards compatibility. It is equivalent to /readonly except that in
473 ACLs where /readonly is required you may specify /leaky/noupdate or
474 /strict/noupdate which are treated the same as /readonly.
476 A useful new feature is the /count= option. This is a generalization
477 of the per_byte option, so that you can measure the throughput of other
478 aggregate values. For example, the per_byte option is now equivalent
479 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
481 The per_rcpt option has been generalized using the /count= mechanism
482 (though it's more complicated than the per_byte equivalence). When it is
483 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
484 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
485 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
486 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
487 Note that using per_rcpt with a non-readonly update mode in more than
488 one ACL will cause the recipients to be double-counted. (The per_mail
489 and per_byte options don't have this problem.)
491 The handling of very low rates has changed slightly. If the computed rate
492 is less than the event's count (usually one) then this event is the first
493 after a long gap. In this case the rate is set to the same as this event's
494 count, so that the first message of a spam run is counted properly.
496 The major new feature is a mechanism for counting the rate of unique
497 events. The new per_addr option counts the number of different
498 recipients that someone has sent messages to in the last time period. It
499 behaves like per_rcpt if all the recipient addresses are different, but
500 duplicate recipient addresses do not increase the measured rate. Like
501 the /count= option this is a general mechanism, so the per_addr option
502 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
503 example, measure the rate that a client uses different sender addresses
504 with the options per_mail/unique=$sender_address. There are further
505 details in the main documentation.
507 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
509 TF/06 Removed a few PCRE remnants.
511 TF/07 Automatically extract Exim's version number from tags in the git
512 repository when doing development or release builds.
514 PP/02 Raise smtp_cmd_buffer_size to 16kB.
515 Bugzilla 879. Patch from Paul Fisher.
517 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
518 Heavily based on revision 40f9a89a from Simon Arlott's tree.
521 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
523 PP/05 Variable $av_failed, true if the AV scanner deferred.
524 Bugzilla 1078. Patch from John Horne.
526 PP/06 Stop make process more reliably on build failure.
527 Bugzilla 1087. Patch from Heiko Schlittermann.
529 PP/07 Make maildir_use_size_file an _expandable_ boolean.
530 Bugzilla 1089. Patch from Heiko Schlittermann.
532 PP/08 Handle ${run} returning more data than OS pipe buffer size.
533 Bugzilla 1131. Patch from Holger Weiß.
535 PP/09 Handle IPv6 addresses with SPF.
536 Bugzilla 860. Patch from Wolfgang Breyha.
538 PP/10 GnuTLS: support TLS 1.2 & 1.1.
540 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
543 PP/11 match_* no longer expand right-hand-side by default.
544 New compile-time build option, EXPAND_LISTMATCH_RHS.
545 New expansion conditions, "inlist", "inlisti".
547 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
549 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
551 PP/14 fix log_write() format string regression from TF/03.
552 Bugzilla 1152. Patch from Dmitry Isaikin.
558 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
560 PP/02 Harmonised TLS library version reporting; only show if debugging.
561 Layout now matches that introduced for other libraries in 4.74 PP/03.
563 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
565 PP/04 New "dns_use_edns0" global option.
567 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
570 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
571 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
573 TK/01 Updated PolarSSL code to 0.14.2.
574 Bugzilla 1097. Patch from Andreas Metzler.
576 PP/07 Catch divide-by-zero in ${eval:...}.
579 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
582 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
583 format-string attack -- SECURITY: remote arbitrary code execution.
585 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
586 time unintentionally subject to list matching rules, letting the header
587 cause arbitrary Exim lookups (of items which can occur in lists, *not*
588 arbitrary string expansion). This allowed for information disclosure.
590 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
591 INT_MIN/-1 -- value coerced to INT_MAX.
597 NM/01 Workround for PCRE version dependency in version reporting
600 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
601 This fixes portability to compilers other than gcc, notably
602 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
604 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
605 makefiles for portability to HP-UX and POSIX correctness.
607 PP/01 Permit LOOKUP_foo enabling on the make command-line.
608 Also via indented variable definition in the Makefile.
609 (Debugging by Oliver Heesakkers).
611 PP/02 Restore caching of spamd results with expanded spamd_address.
612 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
614 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
615 Improves build reliability. Fix from: Frank Elsner
617 NM/02 Fix wide character breakage in the rfc2047 coding
618 Fixes bug 1064. Patch from Andrey N. Oktyabrski
620 NM/03 Allow underscore in dnslist lookups
621 Fixes bug 1026. Patch from Graeme Fowler
623 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
624 Code patches from Adam Ciarcinski of NetBSD.
626 NM/04 Fixed exiqgrep to cope with mailq missing size issue
629 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
630 is logged, to avoid truncation. Patch from John Horne.
632 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
633 Patch from Jakob Hirsch.
635 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
636 SQL string expansion failure details.
637 Patch from Andrey Oktyabrski.
639 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
640 Patch from Simon Arlott.
642 PP/09 New lookups functionality failed to compile on old gcc which rejects
643 extern declarations in function scope.
644 Patch from Oliver Fleischmann
646 PP/10 Use sig_atomic_t for flags set from signal handlers.
647 Check getgroups() return and improve debugging.
648 Fixed developed for diagnosis in bug 927 (which turned out to be
651 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
652 Patch from Mark Zealey.
654 PP/12 Bugzilla 1056: Improved spamd server selection.
655 Patch from Mark Zealey.
657 PP/13 Bugzilla 1086: Deal with maildir quota file races.
658 Based on patch from Heiko Schlittermann.
660 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
661 Patch from Uwe Doering, sign-off by Michael Haardt.
663 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
664 variable declaration deep within a block. Bug and patch from
667 PP/15 lookups-Makefile IRIX compatibilty coercion.
669 PP/16 Make DISABLE_DKIM build knob functional.
671 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
672 Patch by Simon Arlott
674 TF/03 Fix valgrind.h portability to C89 compilers that do not support
675 variable argument macros. Our copy now differs from upstream.
681 TF/01 Failure to get a lock on a hints database can have serious
682 consequences so log it to the panic log.
684 TF/02 Log LMTP confirmation messages in the same way as SMTP,
685 controlled using the smtp_confirmation log selector.
687 TF/03 Include the error message when we fail to unlink a spool file.
689 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
690 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
691 for maintaining out-of-tree patches for some time.
693 PP/01 Bugzilla 139: Documentation and portability issues.
694 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
695 Handle per-OS dynamic-module compilation flags.
697 PP/02 Let /dev/null have normal permissions.
698 The 4.73 fixes were a little too stringent and complained about the
699 permissions on /dev/null. Exempt it from some checks.
700 Reported by Andreas M. Kirchwitz.
702 PP/03 Report version information for many libraries, including
703 Exim version information for dynamically loaded libraries. Created
704 version.h, now support a version extension string for distributors
705 who patch heavily. Dynamic module ABI change.
707 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
708 privilege escalation vulnerability whereby the Exim run-time user
709 can cause root to append content of the attacker's choosing to
712 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
715 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
716 If dropping privileges for untrusted macros, we disabled normal logging
717 on the basis that it would fail; for the Exim run-time user, this is not
718 the case, and it resulted in successful deliveries going unlogged.
719 Fixed. Reported by Andreas Metzler.
725 PP/01 Date: & Message-Id: revert to normally being appended to a message,
726 only prepend for the Resent-* case. Fixes regression introduced in
727 Exim 4.70 by NM/22 for Bugzilla 607.
729 PP/02 Include check_rfc2047_length in configure.default because we're seeing
730 increasing numbers of administrators be bitten by this.
732 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
734 PP/03 Bugzilla 994: added openssl_options main configuration option.
736 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
738 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
740 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
742 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
743 without a peer certificate, leading to a segfault because of an
744 assumption that peers always have certificates. Be a little more
745 paranoid. Problem reported by Martin Tscholak.
747 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
748 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
749 NB: ClamAV planning to remove STREAM in "middle of 2010".
750 CL also introduces -bmalware, various -d+acl logging additions and
751 more caution in buffer sizes.
753 PP/09 Implemented reverse_ip expansion operator.
755 PP/10 Bugzilla 937: provide a "debug" ACL control.
757 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
759 PP/12 Bugzilla 973: Implement --version.
761 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
763 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
765 PP/15 Bugzilla 816: support multiple condition rules on Routers.
767 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
768 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
769 ignore trailing whitespace.
771 JJ/02 prevent non-panic DKIM error from being sent to paniclog
773 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
776 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
777 Notification from Dr Andrew Aitchison.
779 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
780 ExtendedDetectionInfo response format.
781 Notification from John Horne.
783 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
786 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
787 XSL and documented dependency on system catalogs, with examples of how
790 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
793 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
794 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
795 configuration file which is writeable by the Exim user or group.
797 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
798 of configuration files to cover files specified with the -C option if
799 they are going to be used with root privileges, not just the default
802 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
803 option (effectively making it always true).
805 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
806 files to be used while preserving root privileges.
808 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
809 that rogue child processes cannot use them.
811 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
812 run-time user, instead of root.
814 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
815 Exim run-time user without dropping privileges.
817 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
818 result string, instead of calling string_vformat() twice with the same
821 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
822 for other users. Others should always drop root privileges if they use
823 -C on the command line, even for a whitelisted configure file.
825 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
827 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
833 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
834 $data_path, and $header_path variables; fixed documentation bugs and
837 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
838 exipick to access non-standard spools, including the "frozen" queue
841 NM/01 Bugzilla 965: Support mysql stored procedures.
842 Patch from Alain Williams
844 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
846 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
847 Patch from Andreas Metzler
849 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
850 Patch from Kirill Miazine
852 NM/05 Bugzilla 671: Added umask to procmail example.
854 JJ/03 installed exipick 20100323.0, fixing doc bug
856 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
857 directory. Notification and patch from Dan Rosenberg.
859 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
861 TK/02 Improve log output when DKIM signing operation fails.
863 MH/01 Treat the transport option dkim_domain as a colon separated
864 list, not as a single string, and sign the message with each element,
865 omitting multiple occurences of the same signer.
867 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
868 Bugzilla 985, 986. Patch by Simon Arlott
870 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
871 Patch by Simon Arlott
873 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
874 MBX locking. Notification from Dan Rosenberg.
880 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
882 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
884 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
886 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
888 NM/04 Bugzilla 915: Flag broken perl installation during build.
894 TK/01 Added patch by Johannes Berg that expands the main option
895 "spamd_address" if it starts with a dollar sign.
897 TK/02 Write list of recipients to X-Envelope-Sender header when building
898 the mbox-format spool file for content scanning (suggested by Jakob
901 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
902 (http://www.dcc-servers.net/) support via dccifd. Activated by
903 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
905 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
906 by Mark Daniel Reidel <mr@df.eu>.
908 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
909 When building exim an external PCRE library is now needed -
910 PCRE is a system library on the majority of modern systems.
911 See entry on PCRE_LIBS in EDITME file.
913 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
914 conversation. Added nologin parameter to request.
915 Patch contributed by Kirill Miazine.
917 TF/01 Do not log submission mode rewrites if they do not change the address.
919 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
921 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
922 log files in place. Contributed by Roberto Lima.
924 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
926 TF/03 Bugzilla 615: When checking the local_parts router precondition
927 after a local_part_suffix or local_part_prefix option, Exim now
928 does not use the address's named list lookup cache, since this
929 contains cached lookups for the whole local part.
931 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
932 Robert Millan. Documentation is in experimental-spec.txt.
934 TF/04 Bugzilla 668: Fix parallel build (make -j).
936 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
938 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
939 Patch provided by Jan Srzednicki.
941 TF/05 Leading white space used to be stripped from $spam_report which
942 wrecked the formatting. Now it is preserved.
944 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
945 that they are available at delivery time.
947 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
949 TF/08 TLS error reporting now respects the incoming_interface and
950 incoming_port log selectors.
952 TF/09 Produce a more useful error message if an SMTP transport's hosts
953 setting expands to an empty string.
955 NM/06 Bugzilla 744: EXPN did not work under TLS.
956 Patch provided by Phil Pennock.
958 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
959 Patch provided by Richard Godbee.
961 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
962 acl_smtp_notquit, added index entry.
964 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
965 Patch provided by Eugene Bujak.
967 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
968 accept(). Patch provided by Maxim Dounin.
970 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
971 Patch provided by Phil Pennock.
973 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
975 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
976 Patch provided by Brad "anomie" Jorsch.
978 NM/14 Bugzilla 622: Added timeout setting to transport filter.
979 Patch provided by Dean Brooks.
981 TK/05 Add native DKIM support (does not depend on external libraries).
983 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
984 Patch provided by Graeme Fowler.
986 NM/16 Bugzilla 851: Documentation example syntax fix.
988 NM/17 Changed NOTICE file to remove references to embedded PCRE.
990 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
993 NM/19 Bugzilla 745: TLS version reporting.
994 Patch provided by Phil Pennock.
996 NM/20 Bugzilla 167: bool: condition support.
997 Patch provided by Phil Pennock.
999 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1000 clients. Patch provided by Phil Pennock.
1002 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1003 Patch provided by Brad "anomie" Jorsch.
1005 NM/23 Bugzilla 687: Fix misparses in eximstats.
1006 Patch provided by Heiko Schlittermann.
1008 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1009 Patch provided by Heiko Schlittermann.
1011 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1012 plus update to original patch.
1014 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1016 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1017 Patch provided by David Brownlee.
1019 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1021 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1023 NM/30 Bugzilla 888: TLS documentation bugfixes.
1025 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1027 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1028 Unlike the original bugzilla I have changed all shell scripts in src tree.
1030 NM/33 Bugzilla 898: Transport filter timeout fix.
1031 Patch by Todd Rinaldo.
1033 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1034 Patch by Serge Demonchaux.
1036 NM/35 Bugzilla 39: Base64 decode bug fixes.
1037 Patch by Jakob Hirsch.
1039 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1041 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1043 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1045 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1051 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1052 ALT-N's libdkim that I have put here:
1053 http://duncanthrax.net/exim-experimental/
1055 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1056 were called 'true' and it seems that C99 defines that as a reserved
1057 keyword to be used with 'bool' variable types. That means you could
1058 not include C99-style headers which use bools without triggering
1059 build errors in sieve.c.
1061 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1062 as mailq or other aliases. Changed the --help handling significantly
1063 to do whats expected. exim_usage() emits usage/help information.
1065 SC/01 Added the -bylocaldomain option to eximstats.
1067 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1069 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1071 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1077 PH/01 Another patch from the Sieve maintainer.
1079 PH/02 When an IPv6 address is converted to a string for single-key lookup
1080 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1081 dots are used instead of colons so that keys in lsearch files need not
1082 contain colons. This was done some time before quoting was made available
1083 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1084 (notated using the quote facility) so as to distinguish them from IPv4
1085 keys. This meant that lookups for IP addresses in host lists did not work
1086 for iplsearch lookups.
1088 This has been fixed by arranging for IPv6 addresses to be expressed with
1089 colons if the lookup type is iplsearch. This is not incompatible, because
1090 previously such lookups could never work.
1092 The situation is now rather anomolous, since one *can* have colons in
1093 ordinary lsearch keys. However, making the change in all cases is
1094 incompatible and would probably break a number of configurations.
1096 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1099 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1100 conversion specification without a maximum field width, thereby enabling
1101 a rogue spamd server to cause a buffer overflow. While nobody in their
1102 right mind would setup Exim to query an untrusted spamd server, an
1103 attacker that gains access to a server running spamd could potentially
1104 exploit this vulnerability to run arbitrary code as the Exim user.
1106 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1107 $primary_hostname instead of what libspf2 thinks the hosts name is.
1109 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1110 a directory entry by the name of the lookup key. Previously, if a
1111 symlink pointed to a non-existing file or a file in a directory that
1112 Exim lacked permissions to read, a lookup for a key matching that
1113 symlink would fail. Now it is enough that a matching directory entry
1114 exists, symlink or not. (Bugzilla 503.)
1116 PH/03 The body_linecount and body_zerocount variables are now exported in the
1119 PH/04 Added the $dnslist_matched variable.
1121 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1122 This means they are set thereafter only if the connection becomes
1125 PH/06 Added the client_condition to authenticators so that some can be skipped
1126 by clients under certain conditions.
1128 PH/07 The error message for a badly-placed control=no_multiline_responses left
1129 "_responses" off the end of the name.
1131 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1133 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1134 (without spaces) instead of just copying the configuration text.
1136 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1138 PH/11 Added $max_received_linelength.
1140 PH/12 Added +ignore_defer and +include_defer to host lists.
1142 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1143 way in which PCRE > 7.0 is built.
1145 PH/14 Implemented queue_only_load_latch.
1147 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1148 MAIL command. The effect was to mangle the value on 64-bit systems.
1150 PH/16 Another patch from the Sieve maintainer.
1152 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1154 PH/18 If a system quota error occurred while trying to create the file for
1155 a maildir delivery, the message "Mailbox is full" was not appended to the
1156 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1157 only to a quota excession during the actual writing of the file.
1159 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1160 characters?) which causes problems in log lines. The DN values are now
1161 passed through string_printing() before being added to log lines.
1163 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1164 and InterBase are left for another time.)
1166 PH/21 Added message_body_newlines option.
1168 PH/22 Guard against possible overflow in moan_check_errorcopy().
1170 PH/23 POSIX allows open() to be a macro; guard against that.
1172 PH/24 If the recipient of an error message contained an @ in the local part
1173 (suitably quoted, of course), incorrect values were put in $domain and
1174 $local_part during the evaluation of errors_copy.
1180 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1181 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1184 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1185 issue a MAIL command.
1187 PH/02 In an ACL statement such as
1189 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1191 if a client was not listed at all, or was listed with a value other than
1192 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1193 the condition was not true (as it should be), so access was not denied.
1194 The bug was that the ! inversion was incorrectly passed on to the second
1195 item. This has been fixed.
1197 PH/03 Added additional dnslists conditions == and =& which are different from
1198 = and & when the dns lookup returns more than one IP address.
1200 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1201 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1203 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1204 FSYNC, which compiles an option called disable_fsync that allows for
1205 bypassing fsync(). The documentation is heavily laced with warnings.
1207 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1209 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1210 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1211 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1212 including adding "make clean"; (3) Added -fPIC when compiling the test
1213 dynamically loaded module, to get rid of a warning.
1215 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1216 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1217 = 0s. The bug is otherwise harmless.
1219 PH/07 There was a bug in the dovecot authenticator such that the value of
1220 $auth1 could be overwritten, and so not correctly preserved, after a
1221 successful authentication. This usually meant that the value preserved by
1222 the server_setid option was incorrect.
1224 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1226 PH/09 Installed PCRE release 7.0.
1228 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1229 run for batched SMTP input. It is now run at the start of every message
1230 in the batch. While fixing this I discovered that the process information
1231 (output by running exiwhat) was not always getting set for -bs and -bS
1232 input. This is fixed, and it now also says "batched" for BSMTP.
1234 PH/11 Added control=no_pipelining.
1236 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1237 patch, slightly modified), and move the expansion of helo_data till after
1238 the connection is made in the smtp transport (so it can use these
1241 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1243 PH/14 Added log_selector = +pid.
1245 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1247 PH/16 Add ${if forany and ${if forall.
1249 PH/17 Added dsn_from option to vary the From: line in DSNs.
1251 PH/18 Flush SMTP output before performing a callout, unless control =
1252 no_callout_flush is set.
1254 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1255 was true (the default) a successful delivery failed to delete the retry
1256 item, thus causing premature timeout of the address. The bug is now
1259 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1261 PH/21 Long custom messages for fakedefer and fakereject are now split up
1262 into multiline reponses in the same way that messages for "deny" and
1263 other ACL rejections are.
1265 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1266 with slight modification.
1268 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1269 draft, changing the syntax and factoring some duplicate code".
1271 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1272 for deliveries of the second and subsequent messages over the same SMTP
1275 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1276 ${reduce, with only minor "tidies".
1278 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1280 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1281 expansion side effects.
1283 PH/27 When a message times out after an over-quota error from an Exim-imposed
1284 quota, the bounce message says "mailbox is full". This message was not
1285 being given when it was a system quota that was exceeded. It now should
1288 MH/03 Made $recipients available in local_scan(). local_scan() already has
1289 better access to the recipient list through recipients_list[], but
1290 $recipients can be useful in postmaster-provided expansion strings.
1292 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1293 in the case of a MAIL command with additional options following the
1294 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1295 were accidentally chopped off.
1297 PH/29 SMTP synchronization checks are implemented when a command is read -
1298 there is a check that no more input is waiting when there shouldn't be
1299 any. However, for some commands, a delay in an ACL can mean that it is
1300 some time before the response is written. In this time, more input might
1301 arrive, invalidly. So now there are extra checks after an ACL has run for
1302 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1303 pipelining has not been advertised.
1305 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1307 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1308 returned in the SMTP response when smtp_return_error_details was false.
1309 This has been fixed.
1311 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1312 instead of the C I/O that was originally supplied, because problems were
1313 reported on Solaris.
1315 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1316 Exim which did not show up earlier: it was assuming that a call to
1317 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1318 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1319 was a macro that became an assignment, so it seemed to work. This has
1320 changed to a proper function call with a void return, hence the compile
1321 error. Exim's code has been fixed.
1323 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1326 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1328 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1330 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1331 because it turns out to be too big or there is a timeout) there is no
1332 "Completed" line in the log. When some messages of this type were
1333 selected by exigrep, they were listed as "not completed". Others were
1334 picked up by some special patterns. I have improved the selection
1335 criteria to be more general.
1337 PH/38 The host_find_failed option in the manualroute router can now be set
1338 to "ignore", to completely ignore a host whose IP address cannot be
1339 found. If all hosts are ignored, the behaviour is controlled by the new
1340 host_all_ignored option.
1342 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1343 homing or because of multiple MX records with /mx) generated more than
1344 one IP address, and the following item turned out to be the local host,
1345 all the secondary addresses of the first item were incorrectly removed
1346 from the list, along with the local host and any following hosts (which
1347 is what is supposed to happen).
1349 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1350 whoever called Exim into the -H file. In the case of the daemon it was
1351 behaving confusingly. When first started, it used values for whoever
1352 started the daemon, but after a SIGHUP it used the Exim user (because it
1353 calls itself on a restart). I have changed the code so that it now always
1356 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1357 message are rejected with the same error (e.g. no authentication or bad
1358 sender address), and a DATA command is nevertheless sent (as can happen
1359 with PIPELINING or a stupid MUA), the error message that was given to the
1360 RCPT commands is included in the rejection of the DATA command. This is
1361 intended to be helpful for MUAs that show only the final error to their
1364 PH/42 Another patch from the Sieve maintainer.
1366 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1367 Eximstats - Fixed some broken HTML links and added missing column headers
1369 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1370 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1372 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1374 PH/43 Yet another patch from the Sieve maintainer.
1376 PH/44 I found a way to check for a TCP/IP connection going away before sending
1377 the response to the final '.' that terminates a message, but only in the
1378 case where the client has not sent further data following the '.'
1379 (unfortunately, this is allowed). However, in many cases there won't be
1380 any further data because there won't be any more messages to send. A call
1381 to select() can be used: if it shows that the input is "ready", there is
1382 either input waiting, or the socket has been closed. An attempt to read
1383 the next input character can distinguish the two cases. Previously, Exim
1384 would have sent an OK response which the client would never have see.
1385 This could lead to message repetition. This fix should cure that, at
1386 least in a lot of common cases.
1388 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1389 advertised in response to EHLO.
1395 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1396 fixed by 4.65/MH/01 (is this a record?) are fixed:
1398 (i) An empty string was always treated as zero by the numeric comparison
1399 operators. This behaviour has been restored.
1401 (ii) It is documented that the numeric comparison operators always treat
1402 their arguments as decimal numbers. This was broken in that numbers
1403 starting with 0 were being interpreted as octal.
1405 While fixing these problems I realized that there was another issue that
1406 hadn't been noticed. Values of message_size_limit (both the global option
1407 and the transport option) were treated as octal if they started with 0.
1408 The documentation was vague. These values are now always treated as
1409 decimal, and I will make that clear in the documentation.
1415 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1416 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1419 MH/01 Don't check that the operands of numeric comparison operators are
1420 integers when their expansion is in "skipping" mode (fixes bug
1421 introduced by 4.64-PH/07).
1423 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1424 child addresses, Exim now panics and dies. Previously, because the count
1425 is held in a short int, deliveries were likely to be lost. As such a
1426 large number of recipients for a single message is ridiculous
1427 (performance will be very, very poor), I have chosen to impose a limit
1428 rather than extend the field.
1434 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1435 leftover -K file (the existence of which was triggered by #402).
1436 While we were at it, introduced process PID as part of the -K
1437 filename. This should rule out race conditions when creating
1440 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1441 processing considerably. Previous code took too long for large mails,
1442 triggering a timeout which in turn triggers #401.
1444 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1445 in the DK code in transports.c. sendfile() is not really portable,
1446 hence the _LINUX specificness.
1448 TF/01 In the add_headers option to the mail command in an Exim filter,
1449 there was a bug that Exim would claim a syntax error in any
1450 header after the first one which had an odd number of characters
1453 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1454 callout verification, Exim cached a "reject" for the entire domain. This
1455 is correct for most verifications, but it is not correct for a recipient
1456 verification with use_sender or use_postmaster set, because in that case
1457 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1458 case of MAIL FROM:<> rejection from other early rejections (e.g.
1459 rejection of HELO). When verifying a recipient using a non-null MAIL
1460 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1461 Whatever the result of the callout, the value of the domain cache is
1462 left unchanged (for any other kind of callout, getting as far as trying
1463 RCPT means that the domain itself is ok).
1465 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
1468 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
1469 manifest itself as EPIPE rather than ECONNECT. When tidying away a
1470 session, the daemon ignores ECONNECT errors and logs others; it now
1471 ignores EPIPE as well.
1473 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
1474 (quoted-printable decoding).
1476 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
1477 later the small subsequent patch to fix an introduced bug.
1479 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
1481 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
1483 PH/08 An error is now given if message_size_limit is specified negative.
1485 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
1486 to be given (somewhat) arbitrary names.
1488 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
1491 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
1492 miscellaneous code fixes
1494 PH/10 Added the log_reject_target ACL modifier to specify where to log
1497 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
1498 hostname. This is wrong, because it relates to the incoming message (and
1499 probably the interface on which it is arriving) and not to the outgoing
1500 callout (which could be using a different interface). This has been
1501 changed to use the value of the helo_data option from the smtp transport
1502 instead - this is what is used when a message is actually being sent. If
1503 there is no remote transport (possible with a router that sets up host
1504 addresses), $smtp_active_hostname is used.
1506 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
1507 tweaks were necessary in order to get it to work (see also 21 below):
1508 (a) The code assumed that strncpy() returns a negative number on buffer
1509 overflow, which isn't the case. Replaced with Exim's string_format()
1511 (b) There were several signed/unsigned issues. I just did the minimum
1512 hacking in of casts. There is scope for a larger refactoring.
1513 (c) The code used strcasecmp() which is not a standard C function.
1514 Replaced with Exim's strcmpic() function.
1515 (d) The code set only $1; it now sets $auth1 as well.
1516 (e) A simple test gave the error "authentication client didn't specify
1517 service in request". It would seem that Dovecot has changed its
1518 interface. Fortunately there's a specification; I followed it and
1519 changed what the client sends and it appears to be working now.
1521 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
1524 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
1525 address A is aliased to B and C, where B exists and C does not. Without
1526 -v the output is "A verified" because verification stops after a
1527 successful redirection if more than one address is generated. However,
1528 with -v the child addresses are also verified. Exim was outputting "A
1529 failed to verify" and then showing the successful verification for C,
1530 with its parentage. It now outputs "B failed to verify", showing B's
1531 parentage before showing the successful verification of C.
1533 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
1534 look up a TXT record in a specific list after matching in a combined
1537 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
1538 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
1539 they consult the DNS. I had assumed they would set it the way they
1540 wanted; and indeed my experiments on Linux seem to show that in some
1541 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
1542 To be on the safe side, however, I have now made the interface to
1543 host_find_byname() similar to host_find_bydns(), with an argument
1544 containing the DNS resolver options. The host_find_byname() function now
1545 sets these options at its start, just as host_find_bydns() does. The smtp
1546 transport options dns_qualify_single and dns_search_parents are passed to
1547 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
1548 of host_find_byname() use the default settings of RES_DEFNAMES
1549 (qualify_single) but not RES_DNSRCH (search_parents).
1551 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
1552 spool_read_header() do less string testing, by means of a preliminary
1553 switch on the second character of optional "-foo" lines. (This is
1554 overdue, caused by the large number of possibilities that now exist.
1555 Originally there were few.) While I was there, I also converted the
1556 str(n)cmp tests so they don't re-test the leading "-" and the first
1557 character, in the hope this might squeeze out yet more improvement.
1559 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
1560 flag allowing group syntax was set by the header_syntax check but not
1561 turned off, possible causing trouble later; (2) The flag was not being
1562 set at all for the header_verify test, causing "group"-style headers to
1563 be rejected. I have now set it in this case, and also caused header_
1564 verify to ignore an empty address taken from a group. While doing this, I
1565 came across some other cases where the code for allowing group syntax
1566 while scanning a header line wasn't quite right (mostly, not resetting
1567 the flag correctly in the right place). These bugs could have caused
1568 trouble for malformed header lines. I hope it is now all correct.
1570 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
1571 with the "reply" argument non-NULL. The code, however (which originally
1572 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
1573 but it didn't always do it. This confused somebody who was copying the
1574 code for some other use. I have removed all the tests.
1576 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
1577 feature that was used to support insecure browsers during the U.S. crypto
1578 embargo. It requires special client support, and Exim is probably the
1579 only MTA that supported it -- and would never use it because real RSA is
1580 always available. This code has been removed, because it had the bad
1581 effect of slowing Exim down by computing (never used) parameters for the
1582 RSA_EXPORT functionality.
1584 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
1585 authenticator to fail if there's a tab character in the incoming data
1586 (there should never be unless someone is messing about, as it's supposed
1587 to be base64-encoded). Also added, on Timo's advice, the "secured" option
1588 if the connection is using TLS or if the remote IP is the same as the
1589 local IP, and the "valid-client-cert option" if a client certificate has
1592 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
1593 authenticators. This can be used for authorization after authentication
1594 succeeds. (In the case of plaintext, it servers for both authentication
1597 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
1598 if any retry times were supplied.
1600 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
1601 connection, where there is no client IP address to check. In this
1602 situation, the verify now always succeeds.
1604 PH/25 Applied John Jetmore's -Mset patch.
1606 PH/26 Added -bem to be like -Mset, but loading a message from a file.
1608 PH/27 In a string expansion for a processed (not raw) header when multiple
1609 headers of the same name were present, leading whitespace was being
1610 removed from all of them, but trailing whitespace was being removed only
1611 from the last one. Now trailing whitespace is removed from each header
1612 before concatenation. Completely empty headers in a concatenation (as
1613 before) are ignored.
1615 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
1616 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
1618 PH/29 [Removed. This was a change that I later backed out, and forgot to
1619 correct the ChangeLog entry (that I had efficiently created) before
1620 committing the later change.]
1622 PH/30 Exim was sometimes attempting to deliver messages that had suffered
1623 address errors (4xx response to RCPT) over the same connection as other
1624 messages routed to the same hosts. Such deliveries are always "forced",
1625 so retry times are not inspected. This resulted in far too many retries
1626 for the affected addresses. The effect occurred only when there were more
1627 hosts than the hosts_max_try setting in the smtp transport when it had
1628 the 4xx errors. Those hosts that it had tried were not added to the list
1629 of hosts for which the message was waiting, so if all were tried, there
1630 was no problem. Two fixes have been applied:
1632 (i) If there are any address or message errors in an SMTP delivery, none
1633 of the hosts (tried or untried) are now added to the list of hosts
1634 for which the message is waiting, so the message should not be a
1635 candidate for sending over the same connection that was used for a
1636 successful delivery of some other message. This seems entirely
1637 reasonable: after all the message is NOT "waiting for some host".
1638 This is so "obvious" that I'm not sure why it wasn't done
1639 previously. Hope I haven't missed anything, but it can't do any
1640 harm, as the worst effect is to miss an optimization.
1642 (ii) If, despite (i), such a delivery is accidentally attempted, the
1643 routing retry time is respected, so at least it doesn't keep
1644 hammering the server.
1646 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
1647 in ${readsocket because some servers need this prod.
1649 PH/32 Added some extra debug output when updating a wait-xxx database.
1651 PH/33 The hint "could be header name not terminated by colon", which has been
1652 given for certain expansion errors for a long time, was not being given
1653 for the ${if def:h_colon_omitted{... case.
1655 PH/34 The spec says: "With one important exception, whenever a domain list is
1656 being scanned, $domain contains the subject domain." There was at least
1657 one case where this was not true.
1659 PH/35 The error "getsockname() failed: connection reset by peer" was being
1660 written to the panic log as well as the main log, but it isn't really
1661 panic-worthy as it just means the connection died rather early on. I have
1662 removed the panic log writing for the ECONNRESET error when getsockname()
1665 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
1666 runs only) independently of the message's sender address. This meant
1667 that, if the 4xx error was in fact related to the sender, a different
1668 message to the same recipient with a different sender could confuse
1669 things. In particualar, this can happen when sending to a greylisting
1670 server, but other circumstances could also provoke similar problems.
1671 I have changed the default so that the retry time for these errors is now
1672 based a combination of the sender and recipient addresses. This change
1673 can be overridden by setting address_retry_include_sender=false in the
1676 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
1677 remote server are returned as part of bounce messages. This was not
1678 happening for LMTP over a pipe (the lmtp transport), but now it is the
1679 same for both kinds of LMTP.
1681 PH/38 Despite being documented as not happening, Exim was rewriting addresses
1682 in header lines that were in fact CNAMEs. This is no longer the case.
1684 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
1685 and queue runs started by the daemon processed all messages. This has
1686 been fixed so that -R and -S can now usefully be given with -q<time>.
1688 PH/40 Import PCRE release 6.7 (fixes some bugs).
1690 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
1692 PH/42 Give an error if -q is specified more than once.
1694 PH/43 Renamed the variables $interface_address and $interface_port as
1695 $received_ip_address and $received_port, to make it clear that these
1696 values apply to message reception, and not to the outgoing interface when
1697 a message is delivered. (The old names remain recognized, of course.)
1699 PH/44 There was no timeout on the connect() call when using a Unix domain
1700 socket in the ${readsocket expansion. There now is.
1702 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
1703 be meaningful with "accept".
1705 SC/01 Eximstats V1.43
1706 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
1708 SC/02 Eximstats V1.44
1709 Use a glob alias rather than an array ref in the generated
1710 parser. This improves both readability and performance.
1712 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
1713 Collect SpamAssassin and rejection statistics.
1714 Don't display local sender or destination tables unless
1715 there is data to show.
1716 Added average volumes into the top table text output.
1718 SC/04 Eximstats V1.46
1719 Collect data on the number of addresses (recipients)
1720 as well as the number of messages.
1722 SC/05 Eximstats V1.47
1723 Added 'Message too big' to the list of mail rejection
1724 reasons (thanks to Marco Gaiarin).
1726 SC/06 Eximstats V1.48
1727 Mainlog lines which have GMT offsets and are too short to
1728 have a flag are now skipped.
1730 SC/07 Eximstats V1.49 (Alain Williams)
1731 Added the -emptyok flag.
1733 SC/08 Eximstats V1.50
1734 Fixes for obtaining the IP address from reject messages.
1736 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
1737 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
1738 whitesspace changes from 4.64-PH/27
1740 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
1743 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
1744 are found, allow negative numbers in numeric criteria)
1746 JJ/06 exipick.20061117.2, added new $message_body_missing variable
1748 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
1749 to match changes made in 4.64-PH/43
1751 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
1753 PH/47 Put in an explicit test for a DNS lookup of an address record where the
1754 "domain" is actually an IP address, and force a failure. This locks out
1755 those revolvers/nameservers that support "A-for-A" lookups, in
1756 contravention of the specifications.
1758 PH/48 When a host name was looked up from an IP address, and the subsequent
1759 forward lookup of the name timed out, the host name was left in
1760 $sender_host_name, contrary to the specification.
1762 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
1763 restricted to single-key lookups, Exim was not diagnosing an error if
1764 * or *@ was used with a query-style lookup.
1766 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
1768 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
1769 long ago, but noone interested enough thought of it. Let's just say that
1770 the "1.1" means that there are some new functions that weren't there at
1771 some point in the past.
1773 PH/51 Error processing for expansion failure of helo_data from an smtp
1774 transport during callout processing was broken.
1776 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
1777 tested/used via the -bh/-bhc/-bs options.
1779 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
1780 bug, fixed in subsequent PCRE releases).
1782 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
1783 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
1785 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
1791 SC/01 Use a glob alias rather than an array ref in eximstats generated
1792 parser. This improves both readability and performance.
1794 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
1795 Don't display local sender or destination tables in eximstats unless
1796 there is data to show.
1797 Added average volumes into the eximstats top table text output.
1799 SC/03 Collect data on the number of addresses (recipients) as well
1800 as the number of messages in eximstats.
1802 TF/01 Correct an error in the documentation for the redirect router. Exim
1803 does (usually) call initgroups() when daemonizing.
1805 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
1806 with consistent privilege compared to when running as a daemon.
1808 TF/03 Note in the spec that $authenticated_id is not set for local
1809 submissions from trusted users.
1811 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
1812 Thanks to Dean Brooks <dean@iglou.com> for the patch.
1814 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
1815 by adding some example configuration directives to the default
1816 configuration file. A little bit of work is required to uncomment the
1817 directives and define how usernames and passwords are checked, but
1818 there is now a framework to start from.
1820 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
1821 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
1822 without this. I don't know how relevant this is to other LDAP libraries.
1824 PH/02 Add the verb name to the "unknown ACL verb" error.
1826 PH/03 Magnus Holmgren's patch for filter_prepend_home.
1828 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
1830 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
1831 directory not expanded when it should be if an expanded home directory
1832 was set for the address (which is overridden by the transport).
1834 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
1837 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
1838 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
1839 because it is too late at that time, and has no effect.
1841 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
1842 security issue with \' (bugzilla #107). I could not use the
1843 PQescapeStringConn() function, because it needs a PGconn value as one of
1846 PH/08 When testing addresses using -bt, indicate those final addresses that
1847 are duplicates that would not cause an additional delivery. At least one
1848 person was confused, thinking that -bt output corresponded to deliveries.
1849 (Suppressing duplicates isn't a good idea as you lose the information
1850 about possibly different redirections that led to the duplicates.)
1852 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
1853 systems where poll() doesn't work, in particular OS X.
1855 PH/10 Added more information to debugging output for retry time not reached.
1857 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
1858 operations in malware.c.
1860 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
1863 PH/13 If write_rejectlog was set false when logging was sent to syslog with
1864 syslog_duplication set false, log lines that would normally be written
1865 both the the main log and to the reject log were not written to syslog at
1868 PH/14 In the default configuration, change the use of "message" in ACL warn
1869 statements to "add_header".
1871 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
1872 not followed by a command (e.g. "seen endif").
1874 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
1875 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
1878 PH/17 Added extra conditions to the default value of delay_warning_condition
1882 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
1883 { match{$h_precedence:}{(?i)bulk|list|junk} } \
1884 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
1887 The Auto-Submitted: and various List- headers are standardised, whereas I
1888 don't think Precedence: ever was.
1890 PH/18 Refactored debugging code in route_finduser() to show more information,
1891 in particular, the error code if getpwnam() issues one.
1893 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
1894 This is apparently needed in addition to the PH/07 change above to avoid
1895 any possible encoding problems.
1897 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
1898 but not after initializing Perl.
1900 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
1901 output them only if debugging. By default they are written stderr,
1902 apparently, which is not desirable.
1904 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
1907 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
1910 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1912 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1913 authenticated or an ident call has been made. Suppress the default
1914 values for $authenticated_id and $authenticated_sender (but permit -oMai
1915 and -oMas) when testing with -bh.
1917 PH/24 Re-jigged the order of the tests in the default configuration so that the
1918 tests for valid domains and recipients precede the DNS black list and CSA
1919 tests, on the grounds that those ones are more expensive.
1921 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1922 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1923 This bug exists in every version of Exim that I still have, right back to
1926 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1927 However, an attempt to turn on case-sensitivity in a regex key by
1928 including (?-i) didn't work because the subject string was already
1929 lowercased, and the effects were non-intuitive. It turns out that a
1930 one-line patch can be used to allow (?-i) to work as expected.
1936 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1937 other effects) broke the use of negated acl sub-conditions.
1939 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1942 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1943 "Deny" causes Exim to reject the incoming connection with a 554 error.
1944 Unfortunately, if there is a major crisis, such as a disk failure,
1945 tcp-wrappers gives "deny", whereas what one would like would be some
1946 kind of temporary error. A kludge has been added to help with this.
1947 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1948 554 error is used if errno is still zero or contains ENOENT (which occurs
1949 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1952 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1954 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1955 errors. Otherwise a message that provokes a temporary error (when other
1956 messages do not) can cause a whole host to time out.
1958 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1959 addresses were routed directly to files or pipes from a redirect router.
1960 File deliveries just didn't batch; pipe deliveries might have suffered
1963 PH/06 A failure to get a lock for a hints database would erroneously always say
1964 "Failed to get write lock", even when it was really a read lock.
1966 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1967 of 0600. This has been changed to use the value of the lockfile_mode
1968 option (which defaults to 0600).
1970 PH/08 Applied small patch from the Sieve maintainer.
1972 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1973 folder from quota calculations, a direct delivery into this folder messed
1974 up the contents of the maildirsize file. This was because the regex was
1975 used only to exclude .Trash (or whatever) when the size of the mailbox
1976 was calculated. There was no check that a delivery was happening into an
1977 excluded directory. This bug has been fixed by ignoring all quota
1978 processing for deliveries into excluded directories.
1980 PH/10 Added the maildirfolder_create_regex option to appendfile.
1986 PH/01 The code for finding all the local interface addresses on a FreeBSD
1987 system running IPv6 was broken. This may well have applied to all BSD
1988 systems, as well as to others that have similar system calls. The broken
1989 code found IPv4 interfaces correctly, but gave incorrect values for the
1990 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1991 that it would not match correctly against @[] and not recognize the IPv6
1994 PH/02 The ipliteral router was not recognizing addresses of the form user@
1995 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1997 PH/03 Added disable_ipv6.
1999 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2000 decoded form, because it is most often used to construct To: headers
2001 lines in autoreplies, and the decoded form may well be syntactically
2002 invalid. However, $reply_address has leading white space removed, and all
2003 newlines turned into spaces so that the autoreply transport does not
2006 PH/05 If group was specified without a user on a router, and no group or user
2007 was specified on a transport, the group from the router was ignored.
2009 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2010 for visible compile-time settings that can be used to change these
2011 numbers, for those that want even more. Backwards compatibility with old
2012 spool files has been maintained. However, going back to a previous Exim
2013 release will lost any variables that are in spool files.
2015 PH/07 Two small changes when running in the test harness: increase delay when
2016 passing a TCP/IP connection to a new process, in case the original
2017 process has to generate a bounce, and remove special handling of
2018 127.0.0.2 (sic), which is no longer necessary.
2020 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2021 be the same on different OS.
2023 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2026 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2027 whether --show-vars was specified or not
2029 JJ/02 exipick: Added support for new ACL variable spool format introduced
2032 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2033 syntactically invalid From: or Reply-to: line, and a filter used this to
2034 generate an autoreply, and therefore failed to obtain an address for the
2035 autoreply, Exim could try to deliver to a non-existent relative file
2036 name, causing unrelated and misleading errors. What now happens is that
2037 it logs this as a hard delivery error, but does not attempt to create a
2040 PH/11 The exinext utility has a -C option for testing purposes, but although
2041 the given file was scanned by exinext itself; it wasn't being passed on
2042 when Exim was called.
2044 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2045 an end-of-file indication when reading a command response.
2047 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2048 compiled. In many other places in Exim, IPv6 addresses are always
2049 recognized, so I have changed this. It also means that IPv4 domain
2050 literals of the form [IPV4:n.n.n.n] are now always recognized.
2052 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2053 used if the router is not running as root, for example, when verifying at
2054 ACL time, or when using -bh. The debugging output from this situation was
2055 non-existent - all you got was a failure to exec. I have made two
2058 (a) Failures to set uid/gid, the current directory, or a process leader
2059 in a subprocess such as that created by queryprogram now generate
2060 suitable debugging ouput when -d is set.
2062 (b) The queryprogram router detects when it is not running as root,
2063 outputs suitable debugging information if -d is set, and then runs
2064 the subprocess without attempting to change uid/gid.
2066 PH/15 Minor change to Makefile for building test_host (undocumented testing
2069 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2070 additional section of a DNS packet that returns MX or SRV records.
2071 Instead, it always explicitly searches for A/AAAA records. This avoids
2072 major problems that occur when a DNS server includes only records of one
2073 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2074 fixed another bug: if SRV records were looked up and the corresponding
2075 address records were *not* found in the additional section, the port
2076 values from the SRV records were lost.
2078 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2079 using the correct key (the original address) when searching the retry
2080 rules in order to find which one to use for generating the retry hint.
2082 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2083 adding the default one. Similarly, if it contains a Reply-To: header, the
2084 errors_reply_to option, if set, is not used.
2086 PH/19 When calculating a retry time, Exim used to measure the "time since
2087 failure" by looking at the "first failed" field in the retry record. Now
2088 it does not use this if it is later than than the arrival time of the
2089 message. Instead it uses the arrival time. This makes for better
2090 behaviour in cases where some deliveries succeed, thus re-setting the
2091 "first failed" field. An example is a quota failure for a huge message
2092 when small messages continue to be delivered. Without this change, the
2093 "time since failure" will always be short, possible causing more frequent
2094 delivery attempts for the huge message than are intended.
2095 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2097 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2098 $1, $2, $3) because the numerical variables can be reset during some
2099 expansion items (e.g. "match"), thereby losing the authentication data.
2101 PH/21 Make -bV show the size of off_t variables so that the test suite can
2102 decide whether to run tests for quotas > 2G.
2104 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2105 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2106 filecount value is greater than 2G or if a quota value is greater than 2G
2107 on a system where the size of off_t is not greater than 4, a panic error
2110 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2111 never match. The debug and -bh output now contains an explicit error
2112 message indicating a malformed IPv4 address or mask.
2114 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2115 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2116 PH/23 above applies.
2118 PH/25 Do not write to syslog when running in the test harness. The only
2119 occasion when this arises is a failure to open the main or panic logs
2120 (for which there is an explicit test).
2122 PH/26 Added the /no_tell option to "control=freeze".
2124 PH/27 If a host name lookup failed very early in a connection, for example, if
2125 the IP address matched host_lookup and the reverse lookup yielded a name
2126 that did not have a forward lookup, an error message of the form "no IP
2127 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2128 could be logged. Now it outputs the IP address instead of "NULL".
2130 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2131 allows the sender and the authenticated sender to be set when
2132 submitting a message from within Exim. Since child_open_exim() is
2133 documented for local_scan(), the new function should be too.
2135 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2136 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2137 results in an empty string is now treated as unset.
2139 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2141 PH/31 Added sender_verify_fail logging option.
2143 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2144 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2145 tidied the source and removed it altogether.
2147 PH/33 When a queue run was abandoned because the load average was too high, a
2148 log line was always written; now it is written only if the queue_run log
2149 selector is set. In addition, the log line for abandonment now contains
2150 information about the queue run such as the pid. This is always present
2151 in "start" and "stop" lines but was omitted from the "abandon" line.
2153 PH/34 Omit spaces between a header name and the colon in the error message that
2154 is given when verify = headers_syntax fails (if there are lots of them,
2155 the message gets confusing).
2157 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2158 names, as there are now some PTR records that contain slashes. This check
2159 is only to protect against broken name servers that fall over on strange
2160 characters, so the fact that it applies to all lookups doesn't matter.
2162 PH/36 Now that the new test suite is complete, we can remove some of the
2163 special code in Exim that was needed for the old test suite. For example,
2164 sorting DNS records because real resolvers return them in an arbitrary
2165 order. The new test suite's fake resolver always returns records in the
2168 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2169 bounces) except when queue_only is set, to avoid logging races between
2170 the different processes.
2172 PH/38 Panic-die if .include specifies a non-absolute path.
2174 PH/39 A tweak to the "H" retry rule from its user.
2176 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2177 a label. They prevented compilation on older perls.
2179 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2180 a warning to be raised on newish perls.
2182 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2183 on queue. Changes to match documented behaviour of showing count of
2184 messages matching specified criteria.
2186 PH/40 Changed the default ident timeout from 30s to 5s.
2188 PH/41 Added support for the use of login_cap features, on those BSD systems
2189 that have them, for controlling the resources used by pipe deliveries.
2191 PH/42 The content-scanning code uses fopen() to create files in which to put
2192 message data. Previously it was not paying any attention to the mode of
2193 the files. Exim runs with umask(0) because the rest of the code creates
2194 files with open(), and sets the required mode explicitly. Thus, these
2195 files were ending up world-writeable. This was not a big issue, because,
2196 being within the spool directory, they were not world-accessible. I have
2197 created a function called modefopen, which takes an additional mode
2198 argument. It sets umask(777), creates the file, chmods it to the required
2199 mode, then resets the umask. All the relevant calls to fopen() in the
2200 content scanning code have been changed to use this function.
2202 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2203 to 24 hours. This avoids potential overflow problems when processing G
2204 and H retry rules. I suspect nobody ever tinkers with this value.
2206 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2208 PH/45 When the plaintext authenticator is running as a client, the server's
2209 challenges are checked to ensure they are valid base64 strings. By
2210 default, the authentication attempt is cancelled if an invalid string is
2211 received. Setting client_ignore_invalid_base64 true ignores these errors.
2212 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2213 they are received. Thus, the responses can be made to depend on the
2214 challenges. If an invalid string is ignored, an empty string is placed in
2217 PH/46 Messages that are created by the autoreply transport now contains a
2218 References: header, in accordance with RFCs 2822 and 3834.
2220 PH/47 Added authenticated_sender_force to the smtp transport.
2222 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2224 PH/49 Installed latest patch from the Sieve maintainer.
2226 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2227 was also set, the appendfile transport was unnecessarily scanning a
2228 directory of message files (e.g. for maildir delivery) to find the count
2229 of files (along with the size), even though it did not need this
2230 information. It now does the scan only if it needs to find either the
2231 size of the count of files.
2233 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2235 PH/52 Two bugs concerned with error handling when the smtp transport is
2238 (i) Exim was not creating retry information for temporary errors given
2239 for individual recipients after the DATA command when the smtp transport
2240 was used in LMTP mode. This meant that they could be retried too
2241 frequently, and not timed out correctly.
2243 (ii) Exim was setting the flag that allows error details to be returned
2244 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2245 recipients that were returned after the DATA command.
2247 PH/53 This is related to PH/52, but is more general: for any failing address,
2248 when detailed error information was permitted to be returned to the
2249 sender, but the error was temporary, then after the final timeout, only
2250 "retry timeout exceeded" was returned. Now it returns the full error as
2251 well as "retry timeout exceeded".
2253 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2254 do this, and (what is worse) MTAs that accept it.
2256 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2257 will now be deprecated.
2259 PH/56 New os.c-cygwin from the Cygwin maintainer.
2261 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2262 formats (previously only available in exim formats via -bpr, -bpru,
2263 and -bpra. Now also available in native and exiqgrep formats)
2265 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2266 with very large, slow to parse queues
2268 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2270 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2272 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2273 responses to authentication challenges, though it was showing the
2274 challenges; (ii) I've removed the CR characters from the debug output for
2277 PH/58 Allow for the insertion of a newline as well as a space when a string
2278 is turned into more than one encoded-word during RFC 2047 encoding. The
2279 Sieve code now uses this.
2281 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2282 data_4xx, lost_connection, tls_required.
2284 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2285 message was being sent as an SMTP response.
2287 PH/61 Add -l and -k options to exicyclog.
2289 PH/62 When verifying, if an address was redirected to one new address, so that
2290 verification continued, and the new address failed or deferred after
2291 having set something in $address_data, the value of $address_data was not
2292 passed back to the ACL. This was different to the case when no
2293 redirection occurred. The value is now passed back in both cases.
2295 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2296 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2297 use login_cap.h, so on its own it isn't the distinguishing feature. The
2298 new name refers directly to the setclassresources() function.
2300 PH/65 Added configuration files for NetBSD3.
2302 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2304 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2305 is preferred over IPv4.
2307 PH/68 The bounce_return_message and bounce_return_body options were not being
2308 honoured for bounces generated during the reception of non-SMTP messages.
2309 In particular, this applied to messages rejected by the ACL. This bug has
2310 been fixed. However, if bounce_return_message is true and bounce_return_
2311 body is false, the headers that are returned for a non-SMTP message
2312 include only those that have been read before the error was detected.
2313 (In the case of an ACL rejection, they have all been read.)
2315 PH/69 The HTML version of the specification is now built in a directory called
2316 spec_html instead of spec.html, because the latter looks like a path with
2317 a MIME-type, and this confuses some software.
2319 PH/70 Catch two compiler warnings in sieve.c.
2321 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2322 function verify_get_ident() calls ip_connect() to connect a socket, but
2323 if the "connect()" function timed out, ip_connect() used to close the
2324 socket. However, verify_get_ident() also closes the socket later, and in
2325 between Exim writes to the log, which may get opened at this point. When
2326 the socket was closed in ip_connect(), the log could get the same file
2327 descriptor number as the socket. This naturally causes chaos. The fix is
2328 not to close the socket in ip_connect(); the socket should be closed by
2329 the function that creates it. There was only one place in the code where
2330 this was missing, in the iplookup router, which I don't think anybody now
2331 uses, but I've fixed it anyway.
2333 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2334 well as to direct DNS lookups. Otherwise the handling of names in host
2335 lists is inconsistent and therefore confusing.
2341 PH/01 Two changes to the default runtime configuration:
2343 (1) Move the checks for relay_from_hosts and authenticated clients from
2344 after to before the (commented out) DNS black list checks.
2346 (2) Add control=submission to the relay_from_hosts and authenticated
2347 clients checks, on the grounds that messages accepted by these
2348 statements are most likely to be submissions.
2350 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2352 (1) Generate an error if the third argument for the ${prvs expansion is
2355 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2358 (3) Reset the variables that are obtained from the first argument of
2359 ${prvscheck and used in the second argument before leaving the code,
2360 because their memory is reclaimed, so using them afterwards may do
2363 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2364 one (it's much easier than Tom thought :-).
2366 (5) Because of (4), we can now allow for the use of $prvscheck_result
2367 inside the third argument.
2369 PH/03 For some reason, the default setting of PATH when running a command from
2370 a pipe transport was just "/usr/bin". I have changed it to
2373 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2374 anything to be listed in the output from -bV.
2376 PH/05 When a filter generated an autoreply, the entire To: header line was
2377 quoted in the delivery log line, like this:
2379 => >A.N.Other <ano@some.domain> <original@ddress> ...
2381 This has been changed so that it extracts the operative address. There
2382 may be more than one such address. If so, they are comma-separated, like
2385 => >ano@some.domain,ona@other.domain <original@ddress> ...
2387 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2388 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2389 not being looked up in the rDNS to get a host name, Exim was showing the
2390 IP address twice in Received: lines, even though the IP addresses were
2391 identical. For example:
2393 Received: from [1.2.3.4] (helo=[1.2.3.4])
2395 However, if the real host name was known, it was omitting the HELO data
2396 if it matched the actual IP address. This has been tidied up so that it
2397 doesn't show the same IP address twice.
2399 PH/07 When both +timestamp and +memory debugging was on, the value given by
2400 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2401 called by the debug printing, thereby overwriting the timestamp buffer.
2402 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2404 PH/08 When the original message was included in an autoreply transport, it
2405 always said "this is a copy of the message, including all the headers",
2406 even if body_only or headers_only was set. It now gives an appropriate
2409 PH/09 Applied a patch from the Sieve maintainer which:
2411 o fixes some comments
2412 o adds the (disabled) notify extension core
2413 o adds some debug output for the result of if/elsif tests
2414 o points to the current vacation draft in the documentation
2415 and documents the missing references header update
2419 o fixes a bug in processing the envelope test (when testing
2420 multiple envelope elements, the last element determinted the
2423 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2424 Electronic Mail") by including:
2426 Auto-submitted: auto-generated
2428 in the messages that it generates (bounce messages and others, such as
2429 warnings). In the case of bounce messages for non-SMTP mesages, there was
2430 also a typo: it was using "Auto_submitted" (underscore instead of
2431 hyphen). Since every message generated by Exim is necessarily in response
2432 to another message, thes have all been changed to:
2434 Auto-Submitted: auto-replied
2436 in accordance with these statements in the RFC:
2438 The auto-replied keyword:
2440 - SHOULD be used on messages sent in direct response to another
2441 message by an automatic process,
2443 - MUST NOT be used on manually-generated messages,
2445 - MAY be used on Delivery Status Notifications (DSNs) and Message
2446 Disposition Notifications (MDNs),
2448 - MUST NOT be used on messages generated by automatic or periodic
2449 processes, except for messages which are automatic responses to
2452 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2453 to the default Received: header definition.
2455 PH/12 Added log selector acl_warn_skipped (default on).
2457 PH/13 After a successful wildlsearch lookup, discard the values of numeric
2458 variables because (a) they are in the wrong storage pool and (b) even if
2459 they were copied, it wouldn't work properly because of the caching.
2461 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2462 checking when decoding. Apparently there are clients that generate
2463 overlong encoded strings. Why am I not surprised?
2465 PH/15 If the first argument of "${if match_address" was not empty, but did not
2466 contain an "@" character, Exim crashed. Now it writes a panic log message
2467 and treats the condition as false.
2469 PH/16 In autoreply, treat an empty string for "once" the same as unset.
2471 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
2472 extension "envelope-auth". The code is finished and in agreement with
2473 other implementations, but there is no documentation so far and in fact,
2474 nobody wrote the draft yet. This extension is currently #undef'ed, thus
2475 not changing the active code.
2477 Print executed "if" and "elsif" statements when debugging is used. This
2478 helps a great deal to understand what a filter does.
2480 Document more things not specified clearly in RFC3028. I had all this
2481 sorted out, when out of a sudden new issues came to my mind. Oops."
2483 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
2486 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
2487 canonical form (as documented). However, after a host name lookup from
2488 the IP address, check_host() was doing a simple string comparison with
2489 addresses acquired from the DNS when checking that the found name did
2490 have the original IP as one of its addresses. Since any found IPv6
2491 addresses are likely to be in abbreviated form, the comparison could
2492 fail. Luckily, there already exists a function for doing the comparison
2493 by converting both addresses to binary, so now that is used instead of
2494 the text comparison.
2496 PH/20 There was another similar case to PH/19, when a complete host name was
2497 given in a host list; looking up its IP address could give an abbreviated
2498 form, whereas the current host's name might or might not be abbreviated.
2499 The same fix has been applied.
2505 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
2506 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
2509 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
2510 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
2512 PH/03 Typo: missing ".o" in src/pcre/Makefile.
2514 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
2515 header line, restrict the check to what is listed in RFCs 2369 and 2929.
2516 Also, for "Auto-Submitted", treat anything other than "no" as
2517 non-personal, in accordance with RFC 3834. (Previously it treated
2518 anything starting "auto-" as non-personal.)
2520 TF/01 The control=submission/name=... option had a problem with syntax
2521 errors if the name included a slash character. The /name= option
2522 now slurps the rest of the string, so it can include any characters
2523 but it must come last in the list of options (after /sender_retain
2526 PH/05 Some modifications to the interface to the fake nameserver for the new
2534 TK/01 Added the "success_on_redirect" address verification option. See
2535 NewStuff for rationale and an example.
2537 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
2539 PH/02 Patch to exigrep to allow it to work on syslog lines.
2541 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
2542 fread() to skip over the body file's header line, because in Cygwin the
2543 header line is locked and is inaccessible.
2545 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
2546 co-exist for some time) to make it clear that it is the Exim ID that is
2547 referenced, not the Message-ID: header line.
2549 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
2550 string_format() function, because snprintf() does not exist on all
2553 PH/06 The use of forbid_filter_existstest now also locks out the use of the
2554 ${stat: expansion item.
2556 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
2557 protocol synchronization error", to keep the pedants happy.
2559 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
2560 well as for IRIX systems, when gcc is being used. See the host.c source
2563 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
2565 PH/10 Named domain lists were not working if used in a queue_smtp_domains
2568 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
2569 transport and to the smtp transport in LMTP mode.
2571 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
2573 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
2574 run a filter in a subprocess. This could lead to confusion in subsequent
2575 lookups in the parent process. There should also be a search_tidyup() at
2576 the end of the subprocess.
2578 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
2579 only if the host matched helo_try_verify_hosts, which caused the
2580 verification to occur when the EHLO/HELO command was issued. The ACL just
2581 tested the remembered result. Now, if a previous verification attempt has
2582 not happened, "verify = helo" does it there and then.
2584 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
2586 TK/03 Fix log output including CR from clamd.
2588 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
2589 exist provoked a memory error which could cause a segfault.
2591 PH/15 Installed PCRE 6.2
2593 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
2595 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
2596 of the problem. Specifically, suggested +O2 rather than +O1 for the
2599 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
2601 PH/20 If a delivery was routed to a non-standard port by means of an SRV
2602 record, the port was not correctly logged when the outgoing_port log
2603 selector was set (it logged the transort's default port).
2605 PH/21 Added support for host-specific ports to manualroute, queryprogram,
2606 fallback_hosts, and "hosts" in the smtp transport.
2608 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
2609 host errors such as "Connection refused".
2611 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
2612 authentication with radiusclient 0.4.9:
2614 - Error returned from rc_read_config was caught wrongly
2615 - Username/password not passed on to radius server due to wrong length.
2617 The presumption is that some radiusclient API changes for 4.51/PH/17
2618 were not taken care of correctly. The code is still untested by me (my
2619 Linux distribution still has 0.3.2 of radiusclient), but it was
2620 contributed by a Radius user.
2622 PH/24 When doing a callout, the value of $domain wasn't set correctly when
2623 expanding the "port" option of the smtp transport.
2625 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
2626 while reading a MIME header. Thanks to Tom Hughes for a patch.
2628 PH/24 Include config.h inside local_scan.h so that configuration settings are
2631 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
2632 that in an ACL for RCPT (for example), you can examine exactly what was
2635 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
2636 commands, but it was not correctly comparing the address with the actual
2637 client host address. Thus, it would show the EHLO address in Received:
2638 header lines when this was not necessary.
2640 PH/27 Added the % operator to ${eval:}.
2642 PH/28 Exim tries to create and chdir to its spool directory when it starts;
2643 it should be ignoring failures (because with -C, for example, it has lost
2644 privilege). It wasn't ignoring creation failures other than "already
2647 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
2650 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
2651 because an input line was too long, either on its own, or by virtue of
2652 too many continuations, the temporary file was not being removed, and the
2653 return code was incorrect.
2655 PH/31 Missing "BOOL" in function definition in filtertest.c.
2657 PH/32 Applied Sieve patches from the maintainer.
2659 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
2661 PH/33 Added "verify = not_blind".
2663 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
2664 Local/Makefile (with some defaults set). These are used in built scripts
2665 such as exicyclog, but they have never been used in the exim_install
2666 script (though there are many overriding facilities there). I have
2667 arranged that the exim_install script now takes note of these two
2670 PH/35 Installed configuration files for Dragonfly.
2672 PH/36 When a locally submitted message by a trusted user did not contain a
2673 From: header, and the sender address was obtained from -f or from an SMTP
2674 MAIL command, and the trusted user did not use -F to supply a sender
2675 name, $originator_name was incorrectly used when constructing a From:
2676 header. Furthermore, $originator_name was used for submission mode
2677 messages from external hosts without From: headers in a similar way,
2678 which is clearly wrong.
2680 PH/37 Added control=suppress_local_fixups.
2682 PH/38 When log_selector = +received_sender was set, and the addition of the
2683 sender made the log line's construction buffer exactly full, or one byte
2684 less than full, an overflow happened when the terminating "\n" was
2687 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
2688 when the result of a list match is failure because a DNS lookup failed.
2690 PH/40 RM_COMMAND is now used in the building process.
2692 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
2693 the "build-* directories that it finds.
2695 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
2696 address in a domain literal was a prefix of an interface address.
2698 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
2699 when verifying a sender address, unless rewrite_headers is false.
2701 PH/44 Wrote a long comment about why errors_to addresses are verified as
2702 recipients, not senders.
2704 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
2705 the ratelimit ACL was added.
2707 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
2709 PH/46 Added extra information about PostgreSQL errors to the error string.
2711 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
2712 avoiding the need to install special zones in a real server. This is
2713 backwards compatible; if it can't find the fake resolver, it drops back.
2714 Thus, both old and new test suites can be run.
2716 TF/02 Added util/ratelimit.pl
2718 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
2719 clock is set back in time.
2721 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
2722 Candler <B.Candler@pobox.com>.
2724 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
2725 OK for addresses that are the result of redirections.
2727 PH/48 A number of further additions for the benefit of the new test suite,
2728 including a fake gethostbyname() that interfaces to the fake DNS resolver
2731 TF/06 The fix for widen_domains has also been applied to qualify_single and
2732 search_parents which are the other dnslookup options that can cause
2735 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
2738 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
2740 TF/07 Exim produced the error message "an SRV record indicated no SMTP
2741 service" if it encountered an MX record with an empty target hostname.
2742 The message is now "an MX or SRV record indicated no SMTP service".
2744 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
2745 if the DNS of the sending site is misconfigured. This is quite a
2746 common situation. This change restores the behaviour of treating a
2747 helo verification defer as a failure.
2749 PH/51 If self=fail was set on a router, the bounce message did not include the
2750 actual error message.
2756 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
2758 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
2759 command itself ended in error, the underlying message about the transport
2760 filter timeout was being overwritten with the pipe command error. Now the
2761 underlying error message should be appended to the second error message.
2763 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
2765 PH/02 Reduce the amount of output that "make" produces by default. Full output
2766 can still be requested.
2768 PH/03 The warning log line about a condition test deferring for a "warn" verb
2769 was being output only once per connection, rather than after each
2770 occurrence (because it was using the same function as for successful
2771 "warn" verbs). This seems wrong, so I have changed it.
2773 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
2774 it should not have, which might have caused a crash in the right
2775 circumstances, but probably never did.
2777 PH/04 Installed a modified version of Tony Finch's patch to make submission
2778 mode fix the return path as well as the Sender: header line, and to
2779 add a /name= option so that you can make the user's friendly name appear
2782 TF/03 Added the control = fakedefer ACL modifier.
2784 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
2785 Mark Lowes for thorough testing.
2787 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
2789 TK/03 Merged latest SRS patch from Miles Wilton.
2791 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
2792 with the definition in sysexits.h (which is #included earlier).
2793 Fortunately, Exim does not actually use EX_OK. The code used to try to
2794 preserve the sysexits.h value, by assumimg that macro definitions were
2795 scanned for macro replacements. I have been disabused of this notion,
2796 so now the code just undefines EX_OK before #including unistd.h.
2798 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
2799 in the smtp transport. When a block could not be written in a single
2800 write() function, the timeout was being re-applied to each part-write.
2801 This seems wrong - if the receiver was accepting one byte at a time it
2802 would take for ever. The timeout is now adjusted when this happens. It
2803 doesn't have to be particularly precise.
2805 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
2806 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
2808 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
2809 without a domain if the check to <postmaster@domain> fails.
2811 SC/01 Eximstats: added -xls and the ability to specify output files
2812 (patch written by Frank Heydlauf).
2814 SC/02 Eximstats: use FileHandles for outputing results.
2816 SC/03 Eximstats: allow any combination of xls, txt, and html output.
2818 SC/04 Eximstats: fixed display of large numbers with -nvr option
2820 SC/05 Eximstats: fixed merging of reports with empty tables.
2822 SC/06 Eximstats: added the -include_original_destination flag
2824 SC/07 Eximstats: removed tabs and trailing whitespace.
2826 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
2828 TK/06 MBOX spool code: Add real "From " MBOX separator line
2829 so the .eml file is really in mbox format (even though
2830 most programs do not really care). Patch from Alex Miller.
2832 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
2833 The latter is generated from $received_to and is only set if the
2834 message has one envelope recipient. SA can use these headers,
2835 obviously out-of-the-box. Patch from Alex Miller.
2837 PH/08 The ${def test on a variable was returning false if the variable's
2838 value was "0", contrary to what the specification has always said!
2839 The result should be true unless the variable is empty.
2841 PH/09 The syntax error of a character other than { following "${if
2842 def:variable_name" (after optional whitespace) was not being diagnosed.
2843 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
2844 accidental colon was present, for example, could give incorrect results.
2846 PH/10 Tidied the code in a number of places where the st_size field of a stat()
2847 result is used (not including appendfile, where other changes are about
2850 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
2851 This involved changing a lot of size variables from int to off_t. It
2852 should work with maildirs and everything.
2854 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
2855 spamd dying while we are connected to it.
2857 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
2860 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
2861 function for BDB 4.3.
2863 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
2865 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
2866 the sources for PCRE have been re-arranged and more files are now
2869 PH/15 The code I had for printing potentially long long variables in PH/11
2870 above was not the best (it lost precision). The length of off_t variables
2871 is now inspected at build time, and an appropriate printing format (%ld
2872 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
2873 to be "long long int" or "long int". This is needed for the internal
2874 formatting function string_vformat().
2876 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
2877 the configuration file to be ":syslog", then the script "guesses" where
2878 the logs files are, rather than using the compiled in default. In our
2879 case the guess is not the same as the compiled default, so the script
2880 suddenly stopped working when I started to use syslog. The patch checks
2881 to see if log_file_path is "". If so, it attempts to read it from exim
2882 with no configuration file to get the compiled in version, before it
2883 falls back to the previous guessing code."
2885 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
2886 implementing BATV in an Exim configuration. See NewStuff for the gory
2889 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
2890 Makefile that are specific to HP-UX.
2892 PH/18 If the "use_postmaster" option was set for a recipient callout together
2893 with the "random" option, the postmaster address was used as the MAIL
2894 FROM address for the random test, but not for the subsequent recipient
2895 test. It is now used for both.
2897 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
2898 patch removes a few documentation additions to RFC 3028, because the
2899 latest draft now contains them. It adds the new en;ascii-case comparator
2900 and a new error check for 8bit text in MIME parts. Comparator and
2901 require names are now matched exactly. I enabled the subaddress
2902 extension, but it is not well tested yet (read: it works for me)."
2904 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
2905 rework some of the code of TK/09 above to avoid the hardwired use of
2906 "%lld" and "long long". Replaced the call to snprintf() with a call to
2909 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
2910 records point to non-existent hosts", "retry timeout exceeded", and
2911 "retry time not reached for any host after a long failure period".
2913 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2914 experimental DomainKeys support:
2916 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2917 (2) On an error such as an illegally used "control", the wrong name for
2918 the control was given.
2920 These problems did NOT occur unless DomainKeys support was compiled.
2922 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2924 PH/24 Added ${if match_ip condition.
2926 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2927 databases so that it will be absolutely obvious if a crash occurs in the
2928 DB library. This is a regular occurrence (often caused by mis-matched
2931 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2932 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2933 picked up on a user's system that detects such things. There doesn't seem
2934 to be a gcc warning option for this - only an attribute that has to be
2935 put on the function's prototype. It seems that in Fedora Core 4 they have
2936 set this on a number of new functions. No doubt there will be more in due
2939 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2940 specify a transport. However, if an address that was verified by such a
2941 router was the subject of a callout, Exim crashed because it tried to
2942 read the rcpt_include_affixes from the non-existent transport. Now it
2943 just assumes that the setting of that option is false. This bug was
2944 introduced by 4.51/PH/31.
2946 PH/28 Changed -d+all to exclude +memory, because that information is very
2947 rarely of interest, but it makes the output a lot bigger. People tend to
2948 do -d+all out of habit.
2950 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2951 code in os-type was giving problems when libc.so lives in lib64, like on
2954 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2955 aren't the modern standard, and it seems that some systems' include files
2956 don't always have them. Exim was already checking for some of the newer
2957 ones like T_AAAA, and defining it itself. I've added checks for all the
2958 record types that Exim uses.
2960 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2961 not automatically generating a new one, as it is supposed to. This
2962 prevented TLS from working. If the file did exist, but contained invalid
2963 data, a new version was generated, as expected. It was only the case of a
2964 non-existent file that was broken.
2966 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2967 with a change in libdomainkeys > 0.64.
2969 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2970 from DNS. If the selector record carries the flag, it now has
2971 precedence over the domain-wide flag.
2973 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2975 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2976 the use of a port name that isn't defined in /etc/services) occurred, the
2977 message was deferred as in a normal delivery, and thus remained on the
2978 spool, instead of being failed because of the mua_wrapper setting. This
2979 is now fixed, and I tidied up some of the mua_wrapper messages at the
2982 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2983 the messages in a hash of arrays rather than using individual hashes.
2984 This is a bit cleaner and results in dramatic memory savings, albeit
2985 at a slight CPU cost.
2987 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2988 as requested by Marc Sherman.
2990 SC/10 Eximstats: added histograms for user specified patterns as requested
2993 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2995 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2996 fopen() in the content-scanning modules that did not already have it.
3002 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3003 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3005 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3007 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3009 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3010 to test Sieve filters that use "vacation".
3012 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3013 that changes the way the GnuTLS parameters are stored in the cache file.
3014 The new format can be generated externally. For backward compatibility,
3015 if the data in the cache doesn't make sense, Exim assumes it has read an
3016 old-format file, and it generates new data and writes a new file. This
3017 means that you can't go back to an older release without removing the
3020 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3021 work if there are any delivery delays because "one_time" forces the
3022 parent to be marked "delivered", so its unseen clone is never tried
3023 again. For this reason, Exim now forbids the simultaneous setting of
3026 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3027 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3028 entry if you want to know the details. The fix, however, neglected to
3029 consider the case when local delivery batching is involved. The test for
3030 "previously delivered" was not happening when checking to see if an
3031 address could be batched with a previous (undelivered) one; under
3032 certain circumstances this could lead to multiple deliveries to the same
3035 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3036 in its include files, and this causes problems building Exim.
3038 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3039 header_syntax) but Exim was just ignoring anything given after a slash.
3040 In particular, this caused confusion with an attempt to use "verify =
3041 reverse_host_lookup/defer_ok". An error is now given when options are
3042 supplied for verify items that do not have them. (Maybe reverse_host_
3043 lookup should have a defer_ok option, but that's a different point.)
3045 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3046 defined by RFC 821) to 2048, because there were problems with some AUTH
3047 commands, and RFC 1869 says the size should be increased for extended
3048 SMTP commands that take arguments.
3050 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3053 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3054 "unknown" error; now it says that the functionality isn't in the binary.
3056 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3057 an address' error message when a string expansion fails (syntax or
3058 whatever). Otherwise the password may appear in the log. Following change
3059 PH/42 below, there is no longer a chance of it appearing in a bounce
3062 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3064 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3065 of its addresses was ever tried. (Bugzilla bug #2.)
3067 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3068 the result incorrectly in the debug output. (It correctly added a newline
3069 to what was transported.)
3071 TF/01 Added $received_time.
3073 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3074 commented out examples of how to interface to a virus scanner and to
3075 SpamAssassin. Also added commented examples of av_scanner and
3076 spamd_address settings.
3078 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3079 and controls are allowed in which ACLs. There were a couple of minor
3080 errors. Some of the entries in the conditions table (which is a table of
3081 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3082 negation of where the condition IS allowed.
3084 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3086 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3087 header file does not have a version number, so I've had to invent a new
3088 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3089 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3090 radiusclient), but it was contributed by a Radius user.
3092 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3093 files or directories, for OpenSSL.
3095 PH/19 When an Exim process that is running as root has to create an Exim log
3096 file, it does so in a subprocess that runs as exim:exim so as to get the
3097 ownership right at creation (otherwise, other Exim processes might see
3098 the file with the wrong ownership). There was no test for failure of this
3099 fork() call, which would lead to the process getting stuck as it waited
3100 for a non-existent subprocess. Forks do occasionally fail when resources
3101 run out. I reviewed all the other calls to fork(); they all seem to check
3104 PH/20 When checking for unexpected SMTP input at connect time (before writing
3105 the banner), Exim was not dealing correctly with a non-positive return
3106 from the read() function. If the client had disconnected by this time,
3107 the result was a log entry for a synchronization error with an empty
3108 string after "input=" when read() returned zero. If read() returned -1
3109 (an event I could not check), uninitialized data bytes were printed.
3110 There were reports of junk text (parts of files, etc) appearing after
3113 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3115 PH/22 Added support for macro redefinition, and (re)definition in between
3116 driver and ACL definitions.
3118 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3119 forgetting to use the resulting value; it was using the unexpanded value.
3121 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3122 hadn't been configured. The fix is from Juergen Kreileder, who
3123 understands it better than I do:
3125 "Here's what I see happening with three configured cyrus_sasl
3126 authenticators configured (plain, login, cram-md5):
3128 On startup auth_cyrus_sasl_init() gets called for each of these.
3129 This means three calls to sasl_listmech() without a specified mech_list.
3130 => SASL tests which mechs of all available mechs actually work
3131 => three warnings about OTP not working
3132 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3134 With the patch, sasl_listmech() also gets called three times. But now
3135 SASL's mech_list option is set to the server_mech specified in the the
3136 authenticator. Or in other words, the answer from sasl_listmech()
3137 gets limited to just the mech you're testing for (which is different
3139 => the return list contains just 'plain' or 'login', 'cram-md5' or
3140 nothing depending on the value of ob->server_mech.
3142 I've just tested the patch: Authentication still works fine,
3143 unavailable mechs specified in the exim configuration are still
3144 caught, and the auth.log warnings about OTP are gone."
3146 PH/25 When debugging is enabled, the contents of the command line are added
3147 to the debugging output, even when log_selector=+arguments is not
3150 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3151 answer is "GNU", and only if the return is "GNU/something" is the answer
3154 PH/27 $acl_verify_message is now set immediately after the failure of a
3155 verification in an ACL, and so is available in subsequent modifiers. In
3156 particular, the message can be preserved by coding like this:
3158 warn !verify = sender
3159 set acl_m0 = $acl_verify_message
3161 Previously, $acl_verify_message was set only while expanding "message"
3162 and "log_message" when a very denied access.
3164 PH/28 Modified OS/os.c-Linux with
3166 -#ifndef OS_LOAD_AVERAGE
3167 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3169 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3170 nomenclature these days.)
3172 PH/29 Installed patch from the Sieve maintainer that adds the options
3173 sieve_useraddress and sieve_subaddress to the redirect router.
3175 PH/30 In these circumstances:
3176 . Two addresses routed to the same list of hosts;
3177 . First host does not offer TLS;
3178 . First host accepts first address;
3179 . First host gives temporary error to second address;
3180 . Second host offers TLS and a TLS session is established;
3181 . Second host accepts second address.
3182 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3183 and peerdn, if requested) that were in fact used only for the second
3186 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3187 attention to any local part prefix or suffix that was matched by the
3188 router that accepted the address. It now behaves in the same way as it
3189 does for delivery: the affixes are removed from the local part unless
3190 rcpt_include_affixes is set on the transport.
3192 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3193 timeout during the DATA phase of an incoming message.
3195 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3196 applied a patch sanctioned by the Sieve maintainer.
3198 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3199 the uid or gid is negative. A case of a negative gid caused this to be
3200 noticed. The fix allows for either to be negative.
3202 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3203 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3206 PH/36 The change PH/12 above was broken. Fixed it.
3208 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3209 the grounds that routing the same address twice would always produce the
3210 same answer. This might have been true once, but it is certainly no
3211 longer true now. Routing a child address may depend on the previous
3212 routing that produced that child. Some complicated redirection strategies
3213 went wrong when messages had multiple recipients, and made Exim's
3214 behaviour dependent on the order in which the addresses were given.
3216 I have moved the duplicate checking until after the routing is complete.
3217 Exim scans the addresses that are assigned to local and remote
3218 transports, and removes any duplicates. This means that more work will be
3219 done, as duplicates will always all be routed, but duplicates are
3220 presumably rare, so I don't expect this is of any significance.
3222 For deliveries to pipes, files, and autoreplies, the duplicate checking
3223 still happens during the routing process, since they are not going to be
3226 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3227 It corrects a timeout issue with spamd. This is Ian's comment: "The
3228 background is that sometimes spamd either never reads data from a
3229 connection it has accepted, or it never writes response data. The exiscan
3230 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3231 blindly assumes that writes won't block so it may never time out."
3233 PH/39 Allow G after quota size as well as K and M.
3235 PH/40 The value set for $authenticated_id in an authenticator may not contain
3236 binary zeroes or newlines because the value is written to log lines and
3237 to spool files. There was no check on this. Now the value is run through
3238 the string_printing() function so that such characters are converted to
3239 printable escape sequences.
3241 PH/41 $message_linecount is a new variable that contains the total number of
3242 lines in the message. Compare $body_linecount, which is the count for the
3245 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3246 bounce and delay warning messages, except in certain special cases, which
3249 (a) An SMTP error message from a remote host;
3250 (b) A message specified in a :fail: redirection;
3251 (c) A message specified in a "fail" command in a system filter;
3252 (d) A message specified in a FAIL return from the queryprogram router;
3253 (e) A message specified by the cannot_route_message router option.
3255 In these cases only, Exim does include the error details in bounce and
3256 warning messages. There are also a few cases where bland messages such
3257 as "unrouteable address" or "local delivery error" are given.
3259 PH/43 $value is now also set for the "else" part of a ${run expansion.
3261 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3262 being worked on, but at least Exim now implements the latest version to
3265 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3266 process to complete was treated as a delivery failure, a timeout while
3267 writing the message to the pipe was logged, but erroneously treated as a
3268 successful delivery. Such timeouts include transport filter timeouts. For
3269 consistency with the overall process timeout, these timeouts are now
3270 treated as errors, giving rise to delivery failures by default. However,
3271 there is now a new Boolean option for the pipe transport called
3272 timeout_defer, which, if set TRUE, converts the failures into defers for
3273 both kinds of timeout. A transport filter timeout is now identified in
3276 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3277 systems where "make" and "gmake" are different, calling "gmake" at top
3278 level broke things. I've arranged for the value of $(MAKE) to be passed
3279 from the Makefile to this script so that it can call the same version of
3283 A note about Exim versions 4.44 and 4.50
3284 ----------------------------------------
3286 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3287 changes of various kinds. As a consequence, a big documentation update was
3288 needed. This delayed the release for rather longer than seemed good, especially
3289 in the light of a couple of (minor) security issues. Therefore, the changes
3290 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3291 release. So 4.44 and 4.50 are in effect two different branches that both start
3294 I have left the 4.50 change log unchanged; it contains all the changes since
3295 4.43. The change log for 4.44 is below; many of its items are identical to
3296 those for 4.50. This seems to be the most sensible way to preserve the
3297 historical information.
3303 1. Minor wording change to the doc/README.SIEVE file.
3305 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3306 computation of the current number of files was incorrect.
3308 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3309 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3312 4. Give more explanation in the error message when the command for a transport
3313 filter fails to execute.
3315 5. There are several places where Exim runs a non-Exim command in a
3316 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3317 was being done only for the command run by the queryprogram router. It is
3318 now done for all such subprocesses. The other cases are: ${run, transport
3319 filters, and the commands run by the lmtp and pipe transports.
3321 6. Added CONFIGURE_GROUP build-time option.
3323 7. Some older OS have a limit of 256 on the maximum number of file
3324 descriptors. Exim was using setrlimit() to set 1000 as a large value
3325 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3326 systems. I've change it so that if it can't get 1000, it tries for 256.
3328 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3329 was an oversight, and furthermore, ever since the addition of extra
3330 controls (e.g. 4.43/32), the checks on when to allow different forms of
3331 "control" were broken. There should now be diagnostics for all cases when a
3332 control that does not make sense is encountered.
3334 9. Added the /retain_sender option to "control=submission".
3336 10. $recipients is now available in the predata ACL (oversight).
3338 11. Tidy the search cache before the fork to do a delivery from a message
3339 received from the command line. Otherwise the child will trigger a lookup
3340 failure and thereby defer the delivery if it tries to use (for example) a
3341 cached ldap connection that the parent has called unbind on.
3343 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3344 of $address_data from the recipient verification was clobbered by the
3345 sender verification.
3347 13. The value of address_data from a sender verification is now available in
3348 $sender_address_data in subsequent conditions in the ACL statement.
3350 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3352 15. Added a new option "connect=<time>" to callout options, to set a different
3355 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3356 was its contents. (It was OK if the option was not defined at all.)
3358 17. A "Completed" log line is now written for messages that are removed from
3359 the spool by the -Mrm option.
3361 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3362 information about exactly what failed.
3364 19. Added -dd to debug only the daemon process.
3366 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3367 handles timeouts, both on the server side and network timeouts. Renamed the
3368 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3370 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3371 It is now set to "smtps".
3373 22. $host_address is now set to the target address during the checking of
3374 ignore_target_hosts.
3376 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3377 being passed; this would have caused $sender_host_name to have been used if
3378 matching the list had actually called for a host name (not very likely,
3379 since this list is usually IP addresses). A host name is now passed as
3382 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3383 code that specifies a non-restarting handler (typically sigaction() in
3384 modern systems) in an attempt to fix a rare and obscure crash bug.
3386 25. Narrowed the window for a race in the daemon that could cause it to ignore
3387 SIGCHLD signals. This is not a major problem, because they are used only to
3388 wake it up if nothing else does.
3390 26. A malformed maildirsize file could cause Exim to calculate negative values
3391 for the mailbox size or file count. Odd effects could occur as a result.
3392 The maildirsize information is now recalculated if the size or filecount
3395 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3396 support for a long time. Removed HAVE_SYS_VFS_H.
3398 28. Installed the latest version of exipick from John Jetmore.
3400 29. In an address list, if the pattern was not a regular expression, an empty
3401 subject address (from a bounce message) matched only if the pattern was an
3402 empty string. Non-empty patterns were not even tested. This was the wrong
3403 because it is perfectly reasonable to use an empty address as part of a
3404 database query. An empty address is now tested by patterns that are
3405 lookups. However, all the other forms of pattern expect the subject to
3406 contain a local part and a domain, and therefore, for them, an empty
3407 address still always fails if the pattern is not itself empty.
3409 30. Exim went into a mad DNS loop when attempting to do a callout where the
3410 host was specified on an smtp transport, and looking it up yielded more
3411 than one IP address.
3413 31. Re-factored the code for checking spool and log partition space into a
3414 function that finds that data and another that does the check. The former
3415 is then used to implement four new variables: $spool_space, $log_space,
3416 $spool_inodes, and $log_inodes.
3418 32. The RFC2047 encoding function was originally intended for short strings
3419 such as real names; it was not keeping to the 75-character limit for
3420 encoded words that the RFC imposes. It now respects the limit, and
3421 generates multiple encoded words if necessary. To be on the safe side, I
3422 have increased the buffer size for the ${rfc2047: expansion operator from
3425 33. It is now permitted to omit both strings after an "if" condition; if the
3426 condition is true, the result is "true". As before, when the second string
3427 is omitted, a false condition yields an empty string. This makes it less
3428 cumbersome to write custom ACL and router conditions.
3430 34. Failure to deliver a bounce message always caused it to be frozen, even if
3431 there was an errors_to setting on the router. The errors_to setting is now
3434 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3435 canonical form (fully expanded) before being placed in
3436 $sender_host_address.
3438 36. The table in the code that translates DNS record types into text (T_A to
3439 "A" for instance) was missing entries for NS and CNAME. It is just possible
3440 that this could have caused confusion if both these types were looked up
3441 for the same domain, because the text type is used as part of Exim's
3442 per-process caching. But the chance of anyone hitting this buglet seems
3445 37. The dnsdb lookup has been extended in a number of ways.
3447 (1) There is a new type, "zns", which walks up the domain tree until it
3448 finds some nameserver records. It should be used with care.
3450 (2) There is a new type, "mxh", which is like "mx" except that it returns
3451 just the host names, not the priorities.
3453 (3) It is now possible to give a list of domains (or IP addresses) to be
3454 looked up. The behaviour when one of the lookups defers can be
3455 controlled by a keyword.
3457 (4) It is now possible to specify the separator character for use when
3458 multiple records are returned.
3460 38. The dnslists ACL condition has been extended: it is now possible to supply
3461 a list of IP addresses and/or domains to be looked up in a particular DNS
3464 39. Added log_selector=+queue_time_overall.
3466 40. When running the queue in the test harness, wait just a tad after forking a
3467 delivery process, to get repeatability of debugging output.
3469 41. Include certificate and key file names in error message when GnuTLS fails
3470 to set them up, because the GnuTLS error message doesn't include the name
3471 of the failing file when there is a problem reading it.
3473 42. Allow both -bf and -bF in the same test run.
3475 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
3477 44. The "Exiscan patch" is now merged into the mainline Exim source.
3479 45. Sometimes the final signoff response after QUIT could fail to get
3480 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3481 < 0 before doing a fflush(). This bug looks as though it goes back to the
3482 introduction of TLS in release 3.20, but "sometimes" must have been rare
3483 because the tests only now provoked it.
3485 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
3486 (this can affect the format of dates).
3488 47. exim_tidydb, when checking for the continued existence of a message for
3489 which it has found a message-specific retry record, was not finding
3490 messages that were in split spool directories. Consequently, it was
3491 deleting retry records that should have stayed in existence.
3493 48. Steve fixed some bugs in eximstats.
3495 49. The SPA authentication driver was not abandoning authentication and moving
3496 on to the next authenticator when an expansion was forced to fail,
3497 contradicting the general specification for all authenticators. Instead it
3498 was generating a temporary error. It now behaves as specified.
3500 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
3501 (the order specifies the preference for clients). The order is now AES256,
3502 AES128, 3DES, ARCFOUR128.
3504 51. Small patch to Sieve code - explicitly set From: when generating an
3507 52. Exim crashed if a remote delivery caused a very long error message to be
3508 recorded - for instance if somebody sent an entire SpamAssassin report back
3509 as a large number of 550 error lines. This bug was coincidentally fixed by
3510 increasing the size of one of Exim's internal buffers (big_buffer) that
3511 happened as part of the Exiscan merge. However, to be on the safe side, I
3512 have made the code more robust (and fixed the comments that describe what
3515 53. Now that there can be additional text after "Completed" in log lines (if
3516 the queue_time_overall log selector is set), a one-byte patch to exigrep
3517 was needed to allow it to recognize "Completed" as not the last thing in
3520 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
3521 patch that reportedly fixes this has been added. I am not expert enough to
3522 create a test for it. This is what the patch creator wrote:
3524 "I found a little strange behaviour of ldap code when working with
3525 Windows 2003 AD Domain, where users was placed in more than one
3526 Organization Units. When I tried to give exim partial DN, the exit code
3527 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
3528 But simultaneously result of request was absolutely normal ldap result,
3529 so I produce this patch..."
3531 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
3532 REFERENCE, so I have modified the code to exclude the patch when that macro
3535 55. Some experimental protocols are using DNS PTR records for new purposes. The
3536 keys for these records are domain names, not reversed IP addresses. The
3537 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3538 leaves it alone. Component reversal etc. now happens only for IP addresses.
3541 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
3543 57. Double the size of the debug message buffer (to 2048) so that more of very
3544 long debug lines gets shown.
3546 58. The exicyclog utility now does better if the number of log files to keep
3547 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
3549 59. Two changes related to the smtp_active_hostname option:
3551 (1) $smtp_active_hostname is now available as a variable.
3552 (2) The default for smtp_banner uses $smtp_active_hostname instead
3553 of $primary_hostname.
3555 60. The host_aton() function is supposed to be passed a string that is known
3556 to be a valid IP address. However, in the case of IPv6 addresses, it was
3557 not checking this. This is a hostage to fortune. Exim now panics and dies
3558 if the condition is not met. A case was found where this could be provoked
3559 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3560 components; fortuitously, this particular loophole had already been fixed
3561 by change 4.50/55 above.
3563 If there are any other similar loopholes, the new check in host_aton()
3564 itself should stop them being exploited. The report I received stated that
3565 data on the command line could provoke the exploit when Exim was running as
3566 exim, but did not say which command line option was involved. All I could
3567 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3568 running as the user.
3571 61. There was a buffer overflow vulnerability in the SPA authentication code
3572 (which came originally from the Samba project). I have added a test to the
3573 spa_base64_to_bits() function which I hope fixes it.
3576 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
3577 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
3579 63. The daemon start-up calls getloadavg() while still root for those OS that
3580 need the first call to be done as root, but it missed one case: when
3581 deliver_queue_load_max is set with deliver_drop_privilege. This is
3582 necessary for the benefit of the queue runner, because there is no re-exec
3583 when deliver_drop_privilege is set.
3585 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
3586 This has been fixed.
3588 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3589 was in use, was not putting the data itself into the right store pool;
3590 consequently, it could be overwritten for a subsequent message in the same
3591 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3594 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
3596 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
3597 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
3598 the function were treating the return as a boolean value, which happened to
3599 work because 0=false and not-0=true, but is not correct code.
3601 68. The host_aton() function was not handling scoped IPv6 addresses (those
3602 with, for example, "%eth0" on the end) correctly.
3604 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
3605 negated items (that is, ~something) in unsigned ints. Some compilers
3606 apparently mutter when there is no cast.
3608 70. If an address verification called from an ACL failed, and did not produce a
3609 user-specific message (i.e. there was only a "system" message), nothing was
3610 put in $acl_verify_message. In this situation, it now puts the system
3613 71. Change 4.23/11 added synchronization checking at the start of an SMTP
3614 session; change 4.31/43 added the unwanted input to the log line - except
3615 that it did not do this in the start of session case. It now does.
3617 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
3618 This is wrong and can cause the other end to generate a synchronization
3619 error if it is another Exim or anything else that does the synchronization
3620 check. A QUIT command is no longer sent after a timeout.
3622 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
3623 during host lookups.
3625 74. The defer_ok option of callout verification was not working if it was used
3626 when verifying addresses in header lines, that is, for this case:
3628 verify = header_sender/callout=defer_ok
3630 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
3631 those file descriptors could be used for SMTP connections. If anything
3632 wrote to stderr (the example that came up was "warn" in embedded Perl), it
3633 could be sent to the SMTP client, causing chaos. The daemon now opens
3634 stdin, stdout, and stderr to /dev/null when it puts itself into the
3637 76. Arrange for output from Perl's "warn" command to be written to Exim's main
3638 log by default. The user can override this with suitable Perl magic.
3640 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
3641 the log message when discard triggers, was not working for the DATA ACL or
3642 for the non-SMTP ACL.
3644 78. Error message wording change in sieve.c.
3646 79. If smtp_accept_max_per_host was set, the number of connections could be
3647 restricted to fewer than expected, because the daemon was trying to set up
3648 a new connection before checking whether the processes handling previous
3649 connections had finished. The check for completed processes is now done
3650 earlier. On busy systems, this bug wouldn't be noticed because something
3651 else would have woken the daemon, and it would have reaped the completed
3654 80. If a message was submitted locally by a user whose login name contained one
3655 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
3656 It caused a spool format error. I have fixed the spool reading code. A
3657 related problem was that the "from" clause in the Received: line became
3658 illegal because of the space(s). It is now covered by ${quote_local_part.
3660 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
3663 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
3665 83. Patch from Sieve maintainer to fix unterminated string problem in
3666 "vacation" handling.
3668 84. Some minor changes to the Linux configuration files to help with other
3669 OS variants using glibc.
3671 85. One more patch for Sieve to update vacation handling to latest spec.
3674 ----------------------------------------------------
3675 See the note above about the 4.44 and 4.50 releases.
3676 ----------------------------------------------------
3682 1. Change 4.43/35 introduced a bug that caused file counts to be
3683 incorrectly computed when quota_filecount was set in an appendfile
3686 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3687 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3690 3. Give more explanation in the error message when the command for a transport
3691 filter fails to execute.
3693 4. There are several places where Exim runs a non-Exim command in a
3694 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3695 was being done only for the command run by the queryprogram router. It is
3696 now done for all such subprocesses. The other cases are: ${run, transport
3697 filters, and the commands run by the lmtp and pipe transports.
3699 5. Some older OS have a limit of 256 on the maximum number of file
3700 descriptors. Exim was using setrlimit() to set 1000 as a large value
3701 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3702 systems. I've change it so that if it can't get 1000, it tries for 256.
3704 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3705 was an oversight, and furthermore, ever since the addition of extra
3706 controls (e.g. 4.43/32), the checks on when to allow different forms of
3707 "control" were broken. There should now be diagnostics for all cases when a
3708 control that does not make sense is encountered.
3710 7. $recipients is now available in the predata ACL (oversight).
3712 8. Tidy the search cache before the fork to do a delivery from a message
3713 received from the command line. Otherwise the child will trigger a lookup
3714 failure and thereby defer the delivery if it tries to use (for example) a
3715 cached ldap connection that the parent has called unbind on.
3717 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3718 of $address_data from the recipient verification was clobbered by the
3719 sender verification.
3721 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3722 was its contents. (It was OK if the option was not defined at all.)
3724 11. A "Completed" log line is now written for messages that are removed from
3725 the spool by the -Mrm option.
3727 12. $host_address is now set to the target address during the checking of
3728 ignore_target_hosts.
3730 13. When checking ignore_target_hosts for an ipliteral router, no host name was
3731 being passed; this would have caused $sender_host_name to have been used if
3732 matching the list had actually called for a host name (not very likely,
3733 since this list is usually IP addresses). A host name is now passed as
3736 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3737 code that specifies a non-restarting handler (typically sigaction() in
3738 modern systems) in an attempt to fix a rare and obscure crash bug.
3740 15. Narrowed the window for a race in the daemon that could cause it to ignore
3741 SIGCHLD signals. This is not a major problem, because they are used only to
3742 wake it up if nothing else does.
3744 16. A malformed maildirsize file could cause Exim to calculate negative values
3745 for the mailbox size or file count. Odd effects could occur as a result.
3746 The maildirsize information is now recalculated if the size or filecount
3749 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3750 support for a long time. Removed HAVE_SYS_VFS_H.
3752 18. Updated exipick to current release from John Jetmore.
3754 19. Allow an empty sender to be matched against a lookup in an address list.
3755 Previously the only cases considered were a regular expression, or an
3758 20. Exim went into a mad DNS lookup loop when doing a callout where the
3759 host was specified on the transport, if the DNS lookup yielded more than
3762 21. The RFC2047 encoding function was originally intended for short strings
3763 such as real names; it was not keeping to the 75-character limit for
3764 encoded words that the RFC imposes. It now respects the limit, and
3765 generates multiple encoded words if necessary. To be on the safe side, I
3766 have increased the buffer size for the ${rfc2047: expansion operator from
3769 22. Failure to deliver a bounce message always caused it to be frozen, even if
3770 there was an errors_to setting on the router. The errors_to setting is now
3773 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3774 canonical form (fully expanded) before being placed in
3775 $sender_host_address.
3777 24. Updated eximstats to version 1.33
3779 25. Include certificate and key file names in error message when GnuTLS fails
3780 to set them up, because the GnuTLS error message doesn't include the name
3781 of the failing file when there is a problem reading it.
3783 26. Expand error message when OpenSSL has problems setting up cert/key files.
3786 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
3787 (this can affect the format of dates).
3789 28. exim_tidydb, when checking for the continued existence of a message for
3790 which it has found a message-specific retry record, was not finding
3791 messages that were in split spool directories. Consequently, it was
3792 deleting retry records that should have stayed in existence.
3794 29. eximstats updated to version 1.35
3795 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
3796 1.35 - bugfix such that pie charts by volume are generated correctly
3798 30. The SPA authentication driver was not abandoning authentication and moving
3799 on to the next authenticator when an expansion was forced to fail,
3800 contradicting the general specification for all authenticators. Instead it
3801 was generating a temporary error. It now behaves as specified.
3803 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
3804 (the order specifies the preference for clients). The order is now AES256,
3805 AES128, 3DES, ARCFOUR128.
3807 31. Small patch to Sieve code - explicitly set From: when generating an
3810 32. Exim crashed if a remote delivery caused a very long error message to be
3811 recorded - for instance if somebody sent an entire SpamAssassin report back
3812 as a large number of 550 error lines. This bug was coincidentally fixed by
3813 increasing the size of one of Exim's internal buffers (big_buffer) that
3814 happened as part of the Exiscan merge. However, to be on the safe side, I
3815 have made the code more robust (and fixed the comments that describe what
3818 33. Some experimental protocols are using DNS PTR records for new purposes. The
3819 keys for these records are domain names, not reversed IP addresses. The
3820 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3821 leaves it alone. Component reversal etc. now happens only for IP addresses.
3824 34. The host_aton() function is supposed to be passed a string that is known
3825 to be a valid IP address. However, in the case of IPv6 addresses, it was
3826 not checking this. This is a hostage to fortune. Exim now panics and dies
3827 if the condition is not met. A case was found where this could be provoked
3828 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3829 components; fortuitously, this particular loophole had already been fixed
3830 by change 4.50/55 or 4.44/33 above.
3832 If there are any other similar loopholes, the new check in host_aton()
3833 itself should stop them being exploited. The report I received stated that
3834 data on the command line could provoke the exploit when Exim was running as
3835 exim, but did not say which command line option was involved. All I could
3836 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3837 running as the user.
3840 35. There was a buffer overflow vulnerability in the SPA authentication code
3841 (which came originally from the Samba project). I have added a test to the
3842 spa_base64_to_bits() function which I hope fixes it.
3845 36. The daemon start-up calls getloadavg() while still root for those OS that
3846 need the first call to be done as root, but it missed one case: when
3847 deliver_queue_load_max is set with deliver_drop_privilege. This is
3848 necessary for the benefit of the queue runner, because there is no re-exec
3849 when deliver_drop_privilege is set.
3851 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3852 was in use, was not putting the data itself into the right store pool;
3853 consequently, it could be overwritten for a subsequent message in the same
3854 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3857 38. Sometimes the final signoff response after QUIT could fail to get
3858 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3859 < 0 before doing a fflush(). This bug looks as though it goes back to the
3860 introduction of TLS in release 3.20, but "sometimes" must have been rare
3861 because the tests only now provoked it.
3867 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
3868 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
3869 Now it returns an integer. A number of calls were still expecting a T/F
3870 return. Fortuitously, in all cases, the tests worked in OK situations,
3871 which is the norm. However, things would have gone wrong on any write
3872 failures on the smtp file descriptor. This function is used when sending
3873 messages over SMTP and also when doing verify callouts.
3875 2. When Exim is called to do synchronous delivery of a locally submitted
3876 message (the -odf or -odi options), it no longer closes stderr before doing
3879 3. Implemented the mua_wrapper option.
3881 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
3883 5. Implemented the functions header_remove(), header_testname(),
3884 header_add_at_position(), and receive_remove_recipient(), and exported them
3887 6. If an ACL "warn" statement specified the addition of headers, Exim already
3888 inserted X-ACL-Warn: at the start if there was no header name. However, it
3889 was not making this test for the second and subsequent header lines if
3890 there were newlines in the string. This meant that an invalid header could
3891 be inserted if Exim was badly configured.
3893 7. Allow an ACL "warn" statement to add header lines at the start or after all
3894 the Received: headers, as well as at the end.
3896 8. Added the rcpt_4xx retry error code.
3898 9. Added postmaster_mailfrom=xxx to callout verification option.
3900 10. Added mailfrom=xxxx to the callout verification option, for verify=
3903 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
3904 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
3906 12. Inserted some casts to stop certain compilers warning when using pointer
3907 differences as field lengths or precisions in printf-type calls (mostly
3908 affecting debugging statements).
3910 13. Added optional readline() support for -be (dynamically loaded).
3912 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
3913 same clock tick as a message's arrival, so that its received time was the
3914 same as the "first fail" time on the retry record, and that message
3915 remained on the queue past the ultimate address timeout, every queue runner
3916 would try a delivery (because it was past the ultimate address timeout) but
3917 after another failure, the ultimate address timeout, which should have then
3918 bounced the address, did not kick in. This was a "< instead of <=" error;
3919 in most cases the first failure would have been in the next clock tick
3920 after the received time, and all would be well.
3922 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
3923 being recognized when the domain list was tested by the match_domain
3924 condition in an expansion string.
3926 16. Added the ${str2b64: operator.
3928 17. Exim was always calling setrlimit() to set a large limit for the number of
3929 processes, without checking whether the existing limit was already
3930 adequate. (It did check for the limit on file descriptors.) Furthermore,
3931 errors from getrlimit() and setrlimit() were being ignored. Now they are
3932 logged to the main and panic logs, but Exim does carry on, to try to do its
3933 job under whatever limits there are.
3935 18. Imported PCRE 5.0.
3937 19. Trivial typo in log message " temporarily refused connection" (the leading
3940 20. If the log selector return_path_on_delivery was set and an address was
3941 redirected to /dev/null, the delivery process crashed because it assumed
3942 that a return path would always be set for a "successful" delivery. In this
3943 case, the whole delivery is bypassed as an optimization, and therefore no
3946 21. Internal re-arrangement: the function for sending a challenge and reading
3947 a response while authentication was assuming a zero-terminated challenge
3948 string. It's now changed to take a pointer and a length, to allow for
3949 binary data in such strings.
3951 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3953 23. Exim was not respecting finduser_retries when seeking the login of the
3954 uid under which it was called; it was always trying 10 times. (The default
3955 setting of finduser_retries is zero.) Also, it was sleeping after the final
3956 failure, which is pointless.
3958 24. Implemented tls_on_connect_ports.
3960 25. Implemented acl_smtp_predata.
3962 26. If the domain in control=submission is set empty, Exim assumes that the
3963 authenticated id is a complete email address when it generates From: or
3964 Sender: header lines.
3966 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3967 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3968 chown and chgrp in /bin and hostname in /usr/bin.
3970 28. Exim was keeping the "process log" file open after each use, just as it
3971 does for the main log. This opens the possibility of it remaining open for
3972 long periods when the USR1 signal hits a daemon. Occasional processlog
3973 errors were reported, that could have been caused by this. Anyway, it seems
3974 much more sensible not to leave this file open at all, so that is what now
3977 29. The long-running daemon process does not normally write to the log once it
3978 has entered its main loop, and it closes the log before doing so. This is
3979 so that log files can straightforwardly be renamed and moved. However,
3980 there are a couple of unusual error situations where the daemon does write
3981 log entries, and I had neglected to close the log afterwards.
3983 30. The text of an SMTP error response that was received during a remote
3984 delivery was being truncated at 512 bytes. This is too short for some of
3985 the long messages that one sometimes sees. I've increased the limit to
3988 31. It is now possible to make retry rules that apply only when a message has a
3989 specific sender, in particular, an empty sender.
3991 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3992 it possible to be selective about when SMTP synchronization is enforced.
3994 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3996 32. Implemented hosts_connection_nolog.
3998 33. Added an ACL for QUIT.
4000 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4003 35. Added mailbox_size and mailbox_filecount to appendfile.
4005 36. Added control = no_multiline_responses to ACLs.
4007 37. There was a bug in the logic of the code that waits for the clock to tick
4008 in the case where the clock went backwards by a substantial amount such
4009 that the microsecond fraction of "now" was more than the microsecond
4010 fraction of "then" (but the whole seconds number was less).
4012 38. Added support for the libradius Radius client library this is found on
4013 FreeBSD (previously only the radiusclient library was supported).
4019 1. When certain lookups returned multiple values in the form name=value, the
4020 quoting of the values was not always being done properly. Specifically:
4021 (a) If the value started with a double quote, but contained no whitespace,
4023 (b) If the value contained whitespace other than a space character (i.e.
4024 tabs or newlines or carriage returns) it was not quoted.
4025 This fix has been applied to the mysql and pgsql lookups by writing a
4026 separate quoting function and calling it from the lookup code. The fix
4027 should probably also be applied to nisplus, ibase and oracle lookups, but
4028 since I cannot test any of those, I have not disturbed their existing code.
4030 2. A hit in the callout cache for a specific address caused a log line with no
4031 reason for rejecting RCPT. Now it says "Previous (cached) callout
4032 verification failure".
4034 3. There was an off-by-one bug in the queryprogram router. An over-long
4035 return line was truncated at 256 instead of 255 characters, thereby
4036 overflowing its buffer with the terminating zero. As well as fixing this, I
4037 have increased the buffer size to 1024 (and made a note to document this).
4039 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4040 when Exim is waiting for an SMTP response from a remote server, Exim
4041 restarts its select() call on the socket, thereby resetting its timeout.
4042 This is not a problem when such interrupts are rare. Somebody set up a cron
4043 job to run exiwhat every 2 minutes, which is less than the normal select()
4044 timeout (5 or 10 minutes). This meant that the select() timeout never
4045 kicked in because it was always reset. I have fixed this by comparing the
4046 time when an interrupt arrives with the time at the start of the first call
4047 to select(). If more time than the timeout has elapsed, the interrupt is
4048 treated as a timeout.
4050 5. Some internal re-factoring in preparation for the addition of Sieve
4051 extensions (by MH). In particular, the "personal" test is moved to a
4052 separate function, and given an option for scanning Cc: and Bcc: (which is
4053 not set for Exim filters).
4055 6. When Exim created an email address using the login of the caller as the
4056 local part (e.g. when creating a From: or Sender: header line), it was not
4057 quoting the local part when it contained special characters such as @.
4059 7. Installed new OpenBSD configuration files.
4061 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4062 try to make them clearer.
4064 9. Callout options, other than the timeout value, were being ignored when
4065 verifying sender addresses in header lines. For example, when using
4067 verify = header_sender/callout=no_cache
4069 the cache was (incorrectly) being used.
4071 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4072 only the Cygwin environment.
4074 11. When return_path_on_delivery was set as a log selector, if different remote
4075 addresses in the same message used different return paths and parallel
4076 remote delivery occurred, the wrong values would sometimes be logged.
4077 (Whenever a remote delivery process finished, the return path value from
4078 the most recently started remote delivery process was logged.)
4080 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4081 lines when AUTH and/or TLS are in use. This is the "received protocol"
4082 field. Exim used to use "asmtp" for authenticated SMTP, without any
4083 indication (in the protocol name) for TLS use. Now it follows the RFC and
4084 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4085 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4086 names appear in log lines as well as in Received: header lines.
4088 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4089 extensions, and comparison tests, and to fix some bugs.
4091 14. Changes to the "personal" filter test:
4093 (1) The test was buggy in that it was just doing the equivalent of
4094 "contains" tests on header lines. For example, if a user's address was
4095 anne@some.where, the "personal" test would incorrectly be true for
4097 To: susanne@some.where
4099 This test is now done by extracting each address from the header in turn,
4100 and checking the entire address. Other tests that are part of "personal"
4101 are now done using regular expressions (for example, to check local parts
4102 of addresses in From: header lines).
4104 (2) The list of non-personal local parts in From: addresses has been
4105 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4106 taken from the Sieve specification recommendations.
4108 (3) If the message contains any header line starting with "List-" it is
4109 treated as non-personal.
4111 (4) The test for "circular" in the Subject: header line has been removed
4112 because it now seems ill-conceived.
4114 15. Minor typos in src/EDITME comments corrected.
4116 16. Installed latest exipick from John Jetmore.
4118 17. If headers_add on a router specified a text string that was too long for
4119 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4120 of string_sprintf() is now avoided.
4122 18. $message_body_size was not set (it was always zero) when running the DATA
4123 ACL and the local_scan() function.
4125 19. For the "mail" command in an Exim filter, no default was being set for
4126 the once_repeat time, causing a random time value to be used if "once" was
4127 specified. (If the value happened to be <= 0, no repeat happened.) The
4128 default is now 0s, meaning "never repeat". The "vacation" command was OK
4129 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4130 (I found it when inspecting the code).
4132 20. There is now an overall timeout for performing a callout verification. It
4133 defaults to 4 times the callout timeout, which applies to individual SMTP
4134 commands during the callout. The overall timeout applies when there is more
4135 than one host that can be tried. The timeout is checked before trying the
4136 next host. This prevents very long delays if there are a large number of
4137 hosts and all are timing out (e.g. when the network connections are timing
4138 out). The value of the overall timeout can be changed by specifying an
4139 additional sub-option for "callout", called "maxwait". For example:
4141 verify = sender/callout=5s,maxwait=20s
4143 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4144 to the end before writing, but this should make it even safer).
4146 22. Exim was forgetting that it had advertised PIPELINING for the second and
4147 subsequent messages on an SMTP connection. It was also not resetting its
4148 memory on STARTTLS and an internal HELO.
4150 23. When Exim logs an SMTP synchronization error within a session, it now
4151 records whether PIPELINING has been advertised or not.
4153 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4154 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4155 rather than long int.
4157 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4159 26. Added the never_mail option to autoreply.
4165 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4166 crash if the getsockname() call failed; this can happen if a connection is
4167 closed very soon after it is established. The problem was simply in the
4168 order in which certain operations were done, causing Exim to try to write
4169 to the SMTP stream before it had set up the file descriptor. The bug has
4170 been fixed by making things happen in the correct order.
4176 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4177 before the connection was closed, thus losing the rejection response.
4179 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4180 some early Solaris releases, but causes trouble in current releases where
4181 socklen_t is defined.
4183 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4186 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4189 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4190 that releasing the top of it at the end releases what was used for sub-
4191 expansions (unless the block got too big). However, discard this block if
4192 the first thing is a variable or header, so that we can use its block when
4193 it is dynamic (useful for very large $message_headers, for example).
4195 6. Lookups now cache *every* query, not just the most recent. A new, separate
4196 store pool is used for this. It can be recovered when all lookup caches are
4197 flushed. Lookups now release memory at the end of their result strings.
4198 This has involved some general refactoring of the lookup sources.
4200 7. Some code has been added to the store_xxx() functions to reduce the amount
4201 of flapping under certain conditions.
4203 8. log_incoming_interface used to affect only the <= reception log lines. Now
4204 it causes the local interface and port to be added to several more SMTP log
4205 lines, for example "SMTP connection from", and rejection lines.
4207 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4209 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4211 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4212 could be overwritten at the end of the current message (or the start of a
4213 new message if it was set in a HELO ACL). The value is now preserved for
4214 the duration of the SMTP connection.
4216 12. If a transport had a headers_rewrite setting, and a matching header line
4217 contained an unqualified address, that address was qualified, even if it
4218 did not match any rewriting rules. The underlying bug was that the values
4219 of the flags that permit the existence of unqualified sender and recipient
4220 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4221 non-local messages, and by -bnq for local messages) were not being
4222 preserved with the message after it was received.
4224 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4225 "next input=" as part of the text comprising the host identity instead of
4226 the correct text. The code was using the same buffer for two different
4227 strings. However, depending on which order the printing function evaluated
4228 its arguments, the bug did not always show up. Under Linux, for example, my
4229 test suite worked just fine.
4231 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4232 doesn't work with some older versions of Perl. It has been changed to "my",
4233 which in any case is probably the better facility to use.
4235 15. A really picky compiler found some instances of statements for creating
4236 error messages that either had too many or two few arguments for the format
4239 16. The size of the buffer for calls to the DNS resolver has been increased
4240 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4241 for addresses that have a lot of PTR records. This alleviates a problem; it
4242 does not fully solve it.
4244 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4245 buffer now truncates the list and logs the incident, which is the same
4246 action as happens when Exim is looking up a host name and its aliases.
4247 Previously in this situation something unpredictable would happen;
4248 sometimes it was "internal error: store_reset failed".
4250 18. If a server dropped the connection unexpectedly when an Exim client was
4251 using GnuTLS and trying to read a response, the client delivery process
4252 crashed while trying to generate an error log message.
4254 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4255 string, for example:
4257 warn message = H1: something\nH2: something
4259 the text was added as a single header line from Exim's point of view
4260 though it ended up OK in the delivered message. However, searching for the
4261 second and subsequent header lines using $h_h2: did not work. This has been
4262 fixed. Similarly, if a system filter added multiple headers in this way,
4263 the routers could not see them.
4265 20. Expanded the error message when iplsearch is called with an invalid key to
4266 suggest using net-iplsearch in a host list.
4268 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4269 ACLs are no longer actually imposed (and a message to that effect is
4272 22. If a "gecos" field in a passwd entry contained escaped characters, in
4273 particular, if it contained a \" sequence, Exim got it wrong when building
4274 a From: or a Sender: header from that name. A second bug also caused
4275 incorrect handling when an unquoted " was present following a character
4276 that needed quoting.
4278 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4279 was not being matched caselessly.
4281 24. Arranged for all hyphens in the exim.8 source to be escaped with
4284 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4285 properly. Recipient callout cache records were still being keyed to include
4286 the sender, even when use_sender was set false. This led to far more
4287 callouts that were necessary. The sender is no longer included in the key
4288 when use_sender is false.
4290 26. Added "control = submission" modifier to ACLs.
4292 27. Added the ${base62d: operator to decode base 62 numbers.
4294 28. dnsdb lookups can now access SRV records.
4296 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4297 the configuration file.
4299 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4300 (-v) mode. This makes the output for a verbose queue run more intelligible.
4302 31. Added a use_postmaster feature to recipient callouts.
4304 32. Added the $body_zerocount variable, containing the number of binary zero
4305 bytes in the message body.
4307 33. The time of last modification of the "new" subdirectory is now used as the
4308 "mailbox time last read" when there is a quota error for a maildir
4311 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4313 35. Added +ignore_unknown as a special item in host lists.
4315 36. Code for decoding IPv6 addresses in host lists is now included, even if
4316 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4317 address was recognized as an IP address, but was then not correctly decoded
4318 into binary, causing unexpected and incorrect effects when compared with
4325 1. Very minor rewording of debugging text in manualroute to say "list of
4326 hosts" instead of "hostlist".
4328 2. If verify=header_syntax was set, and a header line with an unqualified
4329 address (no domain) and a large number of spaces between the end of the
4330 name and the colon was received, the reception process suffered a buffer
4331 overflow, and (when I tested it) crashed. This was caused by some obsolete
4332 code that should have been removed. The fix is to remove it!
4334 3. When running in the test harness, delay a bit after writing a bounce
4335 message to get a bit more predictability in the log output.
4337 4. Added a call to search_tidyup() just before forking a reception process. In
4338 theory, someone could use a lookup in the expansion of smtp_accept_max_
4339 per_host which, without the tidyup, could leave open a database connection.
4341 5. Added the variables $recipient_data and $sender_data which get set from a
4342 lookup success in an ACL "recipients" or "senders" condition, or a router
4343 "senders" option, similar to $domain_data and $local_part_data.
4345 6. Moved the writing of debug_print from before to after the "senders" test
4348 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4349 problems for message scanning, either using a data ACL, or using
4350 local_scan() because the Received: header was not generated till after they
4351 were called (in order to set the time as the time of reception completion).
4352 I have revised the way this works. The header is now generated after the
4353 body is received, but before the ACL or local_scan() are called. After they
4354 are run, the timestamp in the header is updated.
4360 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4361 before starting a queue runner without re-exec. This happened only when
4362 deliver_drop_privilege was set or when the Exim user was set to root. The
4363 effect of the bug was that timeouts during subsequent deliveries caused
4364 crashes instead of being properly handled. The handler is now left at its
4365 default (and expected) setting.
4367 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4368 message, again when deliver_drop_privilege is set or Exim is run as root.
4369 The bug described in (1) was not present in this case, but the tidying up
4370 of the other signals was missing. I have made the two cases consistent.
4372 3. The ignore_target_hosts setting on a manualroute router was being ignored
4373 for hosts that were looked up using the /MX notation.
4375 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4378 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4379 operated on the sender address. After changing the $sender_address to <>
4380 for the sender address verify, Exim was re-instated it as the original
4381 (before rewriting) address, but remembering that it had rewritten it, so it
4382 wasn't rewriting it again. This bug also had the effect of breaking the
4383 sender address verification caching when the sender address was rewritten.
4385 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4386 This has been changed so that if the ip literal address matches
4387 ignore_target_hosts, the router declines.
4389 7. Added expansion conditions match_domain, match_address, and match_local_
4390 part (NOT match_host).
4392 8. The placeholder for the Received: header didn't have a length field set.
4394 9. Added code to Exim itself and to exim_lock to test for a specific race
4395 condition that could lead to file corruption when using MBX delivery. The
4396 issue is with the lockfile that is created in /tmp. If this file is removed
4397 after a process has opened it but before that process has acquired a lock,
4398 there is the potential for a second process to recreate the file and also
4399 acquire a lock. This could lead to two Exim processes writing to the file
4400 at the same time. The added code performs the same test as UW imapd; it
4401 checks after acquiring the lock that its file descriptor still refers to
4402 the same named file.
4404 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4405 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4408 11. Added the smtp_active_hostname option. If used, this will typically be made
4409 to depend on the incoming interface address. Because $interface_address is
4410 not set up until the daemon has forked a reception process, error responses
4411 that can happen earlier (such as "too many connections") no longer contain
4414 12. If an expansion in a condition on a "warn" statement fails because a lookup
4415 defers, the "warn" statement is abandoned, and the next ACL statement is
4416 processed. Previously this caused the whole ACL to be aborted.
4418 13. Added the iplsearch lookup type.
4420 14. Added ident_timeout as a log selector.
4422 15. Added tls_certificate_verified as a log selector.
4424 16. Added a global option tls_require_ciphers (compare the smtp transport
4425 option of the same name). This controls incoming TLS connections.
4427 17. I finally figured out how to make tls_require_ciphers do a similar thing
4428 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
4429 before starting the TLS session.
4431 18. Tabs are now shown as \t in -bP output.
4433 19. If the log selector return_path_on_delivery was set, Exim crashed when
4434 bouncing a message because it had too many Received: header lines.
4436 20. If two routers both had headers_remove settings, and the first one included
4437 a superfluous trailing colon, the final name in the first list and the
4438 first name in the second list were incorrectly joined into one item (with a
4439 colon in the middle).
4445 1. Added -C and -D options to the exinext utility, mainly to make it easier
4446 to include in the automated testing, but these could be helpful when
4447 multiple configurations are in use.
4449 2. The exinext utility was not formatting the output nicely when there was
4450 an alternate port involved in the retry record key, nor when there was a
4451 message id as well (for retries that were specific to a specific message
4452 and a specific host). It was also confused by IPv6 addresses, because of
4453 the additional colons they contain. I have fixed the IPv4 problem, and
4454 patched it up to do a reasonable job for IPv6.
4456 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
4457 delivery, the log line now contains "pipelined" if PIPELINING was used.
4459 4. An SMTP transport process used to panic and die if the bind() call to set
4460 an explicit outgoing interface failed. This has been changed; it is now
4461 treated in the same way as a connect() failure.
4463 5. A reference to $sender_host_name in the part of a conditional expansion
4464 that was being skipped was still causing a DNS lookup. This no longer
4467 6. The def: expansion condition was not recognizing references to header lines
4468 that used bh_ and bheader_.
4470 7. Added the _cache feature to named lists.
4472 8. The code for checking quota_filecount in the appendfile transport was
4473 allowing one more file than it should have been.
4475 9. For compatibility with Sendmail, the command line option
4483 and sets the incoming protocol and host name (for trusted callers). The
4484 host name and its colon can be omitted when only the protocol is to be set.
4485 Note the Exim already has two private options, -pd and -ps, that refer to
4486 embedded Perl. It is therefore impossible to set a protocol value of "d" or
4487 "s", but I don't think that's a major issue.
4489 10. A number of refactoring changes to the code, none of which should affect
4492 (a) The number of logging options was getting close to filling up the
4493 32-bit word that was used as a bit map. I have split them into two classes:
4494 those that are passed in the argument to log_write(), and those that are
4495 only ever tested independently outside of that function. These are now in
4496 separate 32-bit words, so there is plenty of room for expansion again.
4497 There is no change in the user interface or the logging behaviour.
4499 (b) When building, for example, log lines, the code previously used a
4500 macro that called string_cat() twice, in order to add two strings. This is
4501 not really sufficiently general. Furthermore, there was one instance where
4502 it was actually wrong because one of the argument was used twice, and in
4503 one call a function was used. (As it happened, calling the function twice
4504 did not affect the overall behaviour.) The macro has been replaced by a
4505 function that can join an arbitrary number of extra strings onto a growing
4508 (c) The code for expansion conditions now uses a table and a binary chop
4509 instead of a serial search (which was left over from when there were very
4510 few conditions). Also, it now recognizes conditions like "pam" even when
4511 the relevant support is not compiled in: a suitably worded error message is
4512 given if an attempt is made to use such a condition.
4514 11. Added ${time_interval:xxxxx}.
4516 12. A bug was causing one of the ddress fields not to be passed back correctly
4517 from remote delivery subprocesses. The field in question was not being
4518 subsequently used, so this caused to problems in practice.
4520 13. Added new log selectors queue_time and deliver_time.
4522 14. Might have fixed a bug in maildirsizefile handling that threw up
4523 "unexpected character" debug warnings, and recalculated the data
4524 unnecessarily. In any case, I expanded the warning message to give more
4527 15. Added the message "Restricted characters in address" to the statements in
4528 the default ACL that block characters like @ and % in local parts.
4530 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
4531 Three changes have been made:
4533 (a) There was a serious bug; a negative response to MAIL caused the whole
4534 recipient domain to be cached as invalid, thereby blocking all messages
4535 to all local parts at the same domain, from all senders. This bug has
4536 been fixed. The domain is no longer cached after a negative response to
4537 MAIL if the sender used is not empty.
4539 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
4542 (c) A new callout option, "use_sender" has been added for people who want
4543 the modified behaviour.
4549 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
4552 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
4553 indeed breaks things for older releases.
4555 3. Added additional logging to the case where there is a problem reading data
4556 from a filter that is running in a subprocess using a pipe, in order to
4557 try to track down a specific problem.
4559 4. Testing facility fudge: when running in the test harness and attempting
4560 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
4561 getting "No route to host". Convert this to a timeout.
4563 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
4566 6. Some OS don't have socklen_t but use size_t instead. This affects the
4567 fifth argument of getsockopt() amongst other things. This is now
4568 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
4569 can be set for individual OS. I have set it for SunOS5, OSF1, and
4570 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
4571 some earlier ones do not.
4573 7. Change 4.30/15 was not doing the test caselessly.
4575 8. The standard form for an IPv6 address literal was being rejected by address
4576 parsing in, for example, MAIL and RCPT commands. An example of this kind of
4577 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4578 this, as well as the form without the "IPv6" on the front (but only when
4579 address literals are enabled, of course).
4581 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
4583 10. Exim crashed if a message with an empty sender address specified by -f
4584 encountered a router with an errors_to setting. This could be provoked only
4585 by a command such as
4589 where an empty string was supplied; "<>" did not hit this bug.
4591 11. Installed PCRE release 4.5.
4593 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
4594 remained set. It is now erased.
4596 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
4597 times from message ids (which are base 36 rather than the normal 62).
4599 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
4600 were being counted as actual protocol errors, and logged if the log
4601 selector +smtp_protocol_error was set. One cannot be perfect in this test,
4602 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
4603 and DATA following a set of rejected RCPTs do not count as protocol errors.
4604 In other words, Exim assumes they were pipelined, though this may not
4605 actually be the case. Of course, in all cases the client gets an
4606 appropriate error code.
4608 15. If a lookup fails in an ACL condition, a message about the failure may
4609 be available; it is used if testing the ACL cannot continue, because most
4610 such messages specify what the cause of the deferral is. However, some
4611 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
4612 that caused an old message to be retained and used if a later statement
4613 caused a defer, replacing the real cause of the deferral.
4615 16. If an IP address had so many PTR records that the DNS lookup buffer
4616 was not large enough to hold them, Exim could crash while trying to process
4617 the truncated data. It now detects and logs this case.
4619 17. Further to 4.21/58, another change has been made: if (and only if) the
4620 first line of a message (the first header line) ends with CRLF, a bare LF
4621 in a subsequent header line has a space inserted after it, so as not to
4622 terminate the header.
4624 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
4625 unnecessarily, used atoi() instead of strtol(), and didn't check the
4626 termination when getting file sizes from file names by regex.
4628 19. Completely re-implemented the support for maildirsize files, in the light
4629 of a number of problems with the previous contributed implementation
4630 (4.30/29). In particular:
4632 . If the quota is zero, the maildirsize file is maintained, but no quota is
4635 . If the maildir directory does not exist, it is created before any attempt
4636 to write a maildirsize file.
4638 . The quota value in the file is just a cache; if the quota is changed in
4639 the transport, the new value overrides.
4641 . A regular expression is available for excluding directories from the
4644 20. The autoreply transport checks the characters in options that define the
4645 message's headers; it allows continued headers, but it was checking with
4646 isspace() after an embedded newline instead of explicitly looking for a
4649 21. If all the "regular" hosts to which an address was routed had passed their
4650 expiry times, and had not reached their retry times, the address was
4651 bounced, even if fallback hosts were defined. Now Exim should go on to try
4654 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
4655 equivalent code in the SMTP transport. Some hosts send humungous responses
4656 to HELO/EHLO, more than 1024 it seems.
4658 23. Refactoring: code in filter.c used (void *) for "any old type" but this
4659 gives compiler warnings in some environments. I've now done it "properly",
4662 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
4663 (because of problems with the built-in one) was declared to return uschar *
4664 instead of char *, causing compiler failure.
4666 25. Fixed a file descriptor leak when processing alias/forward files.
4668 26. Fixed a minor format string issue in dbfn.c.
4670 27. Typo in exim.c: ("dmbnz" for "dbmnz").
4672 28. If a filter file refered to $h_xxx or $message_headers, and the headers
4673 contained RFC 2047 "words", Exim's memory could, under certain conditions,
4676 29. When a sender address is verified, it is cached, to save repeating the test
4677 when there is more than one recipient in a message. However, when the
4678 verification involves a callout, it is possible for different callout
4679 options to be set for different recipients. It is too complicated to keep
4680 track of this in the cache, so now Exim always runs a verification when a
4681 callout is required, relying on the callout cache for the optimization.
4682 The overhead is duplication of the address routing, but this should not be
4685 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
4686 to be verified with callout=postmaster, and the main callout worked but the
4687 postmaster check failed, the verification correctly failed. However, if a
4688 subsequent RCPT command asked for sender verification *without* the
4689 postmaster check, incorrect caching caused this verification also to fail,
4692 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
4693 it was not caching the DNS options (qualify_single, search_parents) that
4694 were used when the lookup failed. A subsequent lookup with different
4695 options therefore always gave the same answer, though there were cases
4696 where it should not have. (Example: a "domains = !$mx_any" option on a
4697 dnslookup router: the "domains" option is always processed without any
4698 widening, but the router might have qualify_single set.) Now Exim uses the
4699 cached value only when the same options are set.
4701 32. Added John Jetmore's "exipick" utility to the distribution.
4703 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
4704 than a timeout (e.g. a certificate is required, and is not provided), an
4705 Exim server now closes the connection immediately. Previously it waited for
4706 the client to close - but if the client is SSL, it seems that they each
4707 wait for each other, leading to a delay before one of them times out.
4709 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
4710 maintained 0.8.x compatibility because I don't think many are using it, and
4711 it is clearly obsolete.
4713 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
4716 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
4717 client certificate was expired. A simple patch fixes this, though I don't
4718 understand the full logic of why the verify callback is called multiple
4721 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
4722 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
4723 which causes problems with some clients (such as the Certicom SSL Plus
4724 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
4725 disables the coutermeasure allowing Eudora to connect."
4727 38. Exim was not checking that a write() to a log file succeeded. This could
4728 lead to Bad Things if a log got too big, in particular if it hit a file
4729 size limit. Exim now panics and dies if it cannot write to a log file, just
4730 as it does if it cannot open a log file.
4732 39. Modified OS/Makefile-Linux so that it now contains
4734 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
4736 The two -D definitions ensure that Exim is compiled with large file
4737 support, which makes it possible to handle log files that are bigger than
4740 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
4741 instance) a domain was checked against a named list that involved a lookup,
4742 causing $domain_data to be set, then another domain was checked against the
4743 same list, then the first domain was re-checked, the value of $domain_data
4744 after the final check could be wrong. In particular, if the second check
4745 failed, it could be set empty. This bug probably also applied to
4748 41. The strip_trailing_dot option was not being applied to the address given
4749 with the -f command-line option.
4751 42. The code for reading a message's header from the spool was incrementing
4752 $received_count, but never initializing it. This meant that the value was
4753 incorrect (doubled) while delivering a message in the same process in which
4754 it was received. In the most common configuration of Exim, this never
4755 happens - a fresh exec is done - but it can happen when
4756 deliver_drop_privilege is set.
4758 43. When Exim logs an SMTP synchronization error - client data sent too soon -
4759 it now includes up to 150 characters of the unexpected data in the log
4762 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
4763 and building data strings. The size of both of these buffers was 10 000
4764 bytes - far larger than anybody would *ever* want, thought I. Needless to
4765 say, somebody hit the limit. I have increased the maximum line length to
4766 20 000 and the maximum data length of concatenated lines to 100 000. I have
4767 also fixed two bugs, because there was no checking on these buffers. Tsk,
4768 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
4769 buffer is too small.
4771 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
4772 lsearch lookups. Now it does.
4774 46. When parsing a route_list item in a manualroute router, a fixed-length
4775 buffer was used for the list of hosts. I made this 1024 bytes long,
4776 thinking that nobody would ever have a list of hosts that long. Wrong.
4777 Somebody had a whole pile of complicated expansion conditions, and the
4778 string was silently truncated, leading to an expansion error. It turns out
4779 that it is easier to change to an unlimited length (owing to other changes
4780 that have happened since this code was originally written) than to build
4781 structure for giving a limitation error. The length of the item that
4782 expands into the list of hosts is now unlimited.
4784 47. The lsearch lookup could not handle data where the length of text line was
4785 more than 4095 characters. Such lines were truncated, leading to shortened
4786 data being returned. It should now handle lines of any length.
4788 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
4789 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
4792 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
4793 replacing strings such as BIN_DIRECTORY in a source file: the replacement
4794 no longer happens in comment lines. A list of replacements is now placed
4795 at the head of all of the source files, except those whose only change is
4796 to replace PERL_COMMAND in the very first #! line.
4798 50. Replaced the slow insertion sort in queue.c, for sorting the list of
4799 messages on the queue, with a bottom-up merge sort, using code contributed
4800 by Michael Haardt. This should make operations like -bp somewhat faster on
4801 large queues. It won't affect queue runners, except when queue_run_in_order
4804 51. Installed eximstats 1.31 in the distribution.
4806 52. Added support for SRV lookups to the dnslookup router.
4808 53. If an ACL referred to $message_body or $message_body_end, the value was not
4809 reset for any messages that followed in the same SMTP session.
4811 54. The store-handling optimization for building very long strings was not
4812 differentiating between the different store pools. I don't think this
4813 actually made any difference in practice, but I've tidied it.
4815 55. While running the routers to verify a sender address, $sender_address
4816 was still set to the sender address. This is wrong, because when routing to
4817 send a bounce to the sender, it would be empty. Therefore, I have changed
4818 it so that, while verifying a sender address, $sender_address is set to <>.
4819 (There is no change to what happens when verifying a recipient address.)
4821 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
4822 target A or AAAA records (if not already returned) without resetting the
4823 qualify_single or search_parents options of the DNS resolver. These are
4824 inappropriate in this case because the targets of MX and SRV records must
4825 be FQDNs. A broken DNS record could cause trouble if it happened to have a
4826 target that, when qualified, matched something in the local domain. These
4827 two options are now turned off when doing these lookups.
4829 57. It seems that at least some releases of Reiserfs (which does not have the
4830 concept of a fixed number of inodes) returns zero and not -1 for the
4831 number of available inodes. This interacted badly with check_spool_inodes,
4832 which assumed that -1 was the "no such thing" setting. What I have done is
4833 to check that the total number of inodes is greater than zero before doing
4834 the test of how many are available.
4836 58. When a "warn" ACL statement has a log_message modifier, the message is
4837 remembered, and not repeated. This is to avoid a lot of repetition when a
4838 message has many recipients that cause the same warning to be written.
4839 Howewer, Exim was preserving the list of already written lines for an
4840 entire SMTP session, which doesn't seem right. The memory is now reset if a
4841 new message is started.
4843 59. The "rewrite" debugging flag was not showing the result of rewriting in the
4844 debugging output unless log_rewrite was also set.
4846 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
4847 of (int)(handle) when we know that handle contains (void *)(-1).
4849 61. The Exim daemon panic-logs an error return when it closes the incoming
4850 connection. However "connection reset by peer" seems to be common, and
4851 isn't really an error worthy of noting specially, so that particular error
4854 62. When Exim is trying to find all the local interfaces, it used to panic and
4855 die if the ioctl to get the interface flags failed. However, it seems that
4856 on at least one OS (Solaris 9) it is possible to have an interface that is
4857 included in the list of interfaces, but for which you get a failure error
4858 for this call. This happens when the interface is not "plumbed" into a
4859 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
4860 failure of the "get flags" call assumes that the interface is down.
4862 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
4863 makes life easier for people who are doing arithmetic on fields extracted
4864 from dates, where you often get leading zeros that should not be
4865 interpreted as octal.
4867 64. Added qualify_domain to the redirect router, to override the global
4870 65. If a pathologically long header line contained very many addresses (the
4871 report of this problem mentioned 10 000) and each of them was rewritten,
4872 Exim could use up a very large amount of memory. (It kept on making new
4873 copies of the header line as it rewrote, and never released the old ones.)
4874 At the expense of a bit more processing, the header rewriting function has
4875 been changed so that it no longer eats memory in this way.
4877 66. The generation of the Received: header has been moved from the time that a
4878 message starts to be received, to the time that it finishes. The timestamp
4879 in the Received: header should now be very close to that of the <= log
4880 line. There are two side-effects of this change:
4882 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
4883 logged header lines no longer include the local Received: line, because
4884 it has not yet been created. The same applies to a copy of the message
4885 that is returned to a non-SMTP sender when a message is rejected.
4887 (b) When a filter file is tested using -bf, no additional Received: header
4888 is added to the test message. After some thought, I decided that this
4891 This change does not affect the value of $received_for. It is still set
4892 after address rewriting, but before local_scan() is called.
4894 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
4896 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
4897 gave an unhelpful panic error message, and a defer error. I have managed to
4898 change this behaviour so that it now rejects any supplied certificate,
4899 which seems right, as the list of acceptable certificates is empty.
4901 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
4902 gave an unhelpful defer error. I have not managed to make this reject any
4903 supplied certificates, but the error message it gives is "no certificate
4904 supplied", which is not helpful.
4906 70. exigrep's output now also includes lines that are not associated with any
4907 message, but which match the given pattern. Implemented by a patch from
4908 Martin Sluka, which also tidied up the Perl a bit.
4910 71. Recipient callout verification, like sender verification, was using <> in
4911 the MAIL FROM command. This isn't really the right thing, since the actual
4912 sender may affect whether the remote host accepts the recipient or not. I
4913 have changed it to use the actual sender in the callout; this means that
4914 the cache record is now keyed on a recipient/sender pair, not just the
4915 recipient address. There doesn't seem to be a real danger of callout loops,
4916 since a callout by the remote host to check the sender would use <>.
4917 [SEE ABOVE: changed after hitting problems.]
4919 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
4920 temporary errors. However, in the case of such a code being given after
4921 the end of a data transmission (i.e. after ".") Exim was failing to write
4922 a retry record for the message. (Yes, there was some broken host that was
4923 actually sending 8xx at this point.)
4925 73. An unknown lookup type in a host list could cause Exim to panic-die when
4926 the list was checked. (An example that provoked this was putting <; in the
4927 middle of a list instead of at the start.) If this happened during a DATA
4928 ACL check, a -D file could be left lying around. This kind of configuration
4929 error no longer causes Exim to die; instead it causes a defer errror. The
4930 incident is still logged to the main and panic logs.
4932 74. Buglet left over from Exim 3 conversion. The message "too many messages
4933 in one connection" was written to the rejectlog but not the mainlog, except
4934 when address rewriting (yes!) was being logged.
4936 75. Added write_rejectlog option.
4938 76. When a system filter was run not as root (that is, when system_filter_user
4939 was set), the values of the $n variables were not being returned to the
4940 main process; thus, they were not subsequently available in the $sn
4943 77. Added +return_path_on_delivery log selector.
4945 78. A connection timeout was being treated differently from recipients deferred
4946 when testing hosts_max_try with a message that was older than the host's
4947 retry timeout. (The host should not be counted, thus allowing all hosts to
4948 be tried at least once before bouncing.) This may have been the cause of an
4949 occasionally reported bug whereby a message would remain on the queue
4950 longer than the retry timeout, but would be bounced if a delivery was
4951 forced. I say "may" because I never totally pinned down the problem;
4952 setting up timeout/retry tests is difficult. See also the next item.
4954 79. The ultimate address timeout was not being applied to errors that involved
4955 a combination of host plus message (for example, a timeout on a MAIL
4956 command). When an address resolved to a number of possible hosts, and they
4957 were not all tried for each delivery (e.g. because of hosts_max_try), a
4958 message could remain on the queue longer than the retry timeout.
4960 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4963 81. Fixed an obscure SMTP outgoing bug which required at least the following
4964 conditions: (a) there was another message waiting for the same server;
4965 (b) the server returned 5xx to all RCPT commands in the first message so
4966 that the message was not completed; (c) the server dropped the connection
4967 or gave a negative response to the RSET that Exim sends to abort the
4968 transaction. The observed case was a dropped connection after DATA that had
4969 been sent in pipelining mode. That is, the server had advertised PIPELINING
4970 but was not implementing it correctly. The effect of the bug was incorrect
4971 behaviour, such as trying another host, and this could lead to a crash.
4977 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4978 and daemon.c were passed as pointers to ints; they should have been
4979 pointers to socklen_t variables (which are typically unsigned ints).
4981 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4984 3. Fixed a really odd bug that affected only the testing scheme; patching a
4985 certain fixed string in the binary changed the value of another string that
4986 happened to be identical to the end of the original first string.
4988 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4989 name", it returns that address as the IP address. On some operating
4990 systems (e.g. Solaris), it also passes back the IP address string as the
4991 "host name". However, on others (e.g. Linux), it passes back an empty
4992 string. Exim wasn't checking for this, and was changing the host name to an
4993 empty string, assuming it had been canonicized.
4995 5. Although rare, it is permitted to have more than one PTR record for a given
4996 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4997 all the names associated with an address, because they do in Solaris.
4998 However, it seems that they do not in Linux for data that comes from the
4999 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5000 I found this out when I moved to a new Linux workstation and tried to run
5001 the Exim test suite.
5003 To get round this problem I have changed the code so that it now does its
5004 own call to the DNS to look up PTR records when searching for a host name.
5005 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5006 addresses that are only in /etc/hosts are still found.
5008 This behaviour is, however, controlled by an option called host_lookup_
5009 order, which defaults to "bydns:byaddr". If people want to use the other
5010 order, or indeed, just use one or the other means of lookup, they can
5011 specify it in this variable.
5013 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5014 some operating systems, this comes back from gethostbyaddr() as an empty
5015 string, and this is what Exim used to test for. However, it seems that in
5016 other systems, "." is yielded. Exim now tests for this case too.
5018 7. The values of check_spool_space and check_log_space are now held internally
5019 as a number of kilobytes instead of an absolute number of bytes. If a
5020 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5021 kilobyte. This means that much larger values can be stored.
5023 8. Exim monitor: an attempt to get the action menu when not actually pointing
5024 at a message produces an empty menu entitled "No message selected". This
5025 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5026 no entries in it ("Shell widget menu has zero width and/or height"). So I
5027 have added a single, blank menu entry in this case.
5029 9. Added ${quote_local_part.
5031 10. MIME decoding is now applied to the contents of Subject: header lines when
5034 11. Now that a reference to $sender_host_address automatically causes a reverse
5035 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5036 host lookup before query-style lookups in lists that might use this
5037 variable. This has therefore been abolished, and the "net-" prefix is no
5038 longer necessary for query-style lookups.
5040 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5041 have been a typo for LFLAGS, so it has been changed.
5043 13. The install script calls Exim with "-C /dev/null" in order to find the
5044 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5045 to be output. Howeve, since Exim outputs its version number before the
5046 error, it didn't break the script. It just looked ugly. I fixed this by
5047 always allowing "-C /dev/null" if the caller is root.
5049 14. Ignore overlarge ACL variable number when reading spool file - insurance
5050 against a later release with more variables having written the file.
5052 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5053 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5054 this, as well as the form without the "IPv6" on the front.
5056 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5057 OS/Makefile-Darwin file.
5059 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5060 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5062 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5063 variable. However, some OS use ints for inodes. I've added cast to long int
5064 to get rid of the compiler warning.
5066 19. I had forgotten to lock out "/../" in configuration file names when
5067 ALT_CONFIG_PREFIX was set.
5069 20. Routers used for verification do not need to specify transports. However,
5070 if such a router generated a host list, and callout was configured, Exim
5071 crashed, because it could not find a port number from the (non-existent)
5072 transport. It now assumes port 25 in this circumstance.
5074 21. Added the -t option to exigrep.
5076 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5077 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5078 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5079 them is set, LOOKUP_LSEARCH is forced.
5081 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5083 24. Added sender and host information to the "rejected by local_scan()" log
5084 line; previously there was no indication of these.
5086 25. Added .include_if_exists.
5088 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5089 Linux. It turns out that not all file systems support this. Apparently some
5090 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5091 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5092 not supported on the file descriptor, is now ignored when Exim is trying to
5093 sync a directory. This applies only to Linux.
5095 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5097 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5098 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5101 29. Added support for maildirsize files from supplied patch (modified a bit).
5103 30. The use of :fail: followed by an empty string could lead Exim to respond to
5104 sender verification failures with (e.g.):
5106 550 Verification failed for <xxx>
5107 550 Sender verify failed
5109 where the first response line was missing the '-' that indicates it is not
5110 the final line of the response.
5112 31. The loop for finding the name of the user that called Exim had a hardwired
5113 limit of 10; it now uses the value of finduser_retries, which is used for
5114 all other user lookups.
5116 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5119 33. Exim was neglecting to zero errno before one call of strtol() when
5120 expanding a string and expecting an integer value. On some systems this
5121 resulted in spurious "integer overflow" errors. Also, it was casting the
5122 result into an int without checking.
5124 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5125 did not work. The code looks as if it has *never* worked, though it appears
5126 to have been documented since at least releast 1.62. I have made it work.
5128 35. The "timeout_DNS" error in retry rules, also documented since at least
5129 1.62, also never worked. As it isn't clear exactly what this means, and
5130 clearly it isn't a major issue, I have abolished the feature by treating it
5131 as "timeout", and writing a warning to the main and panic logs.
5133 36. The display of retry rules for -brt wasn't always showing the error code
5136 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5137 timeout_connect_A, timeout_connect_MX.
5139 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5140 to the empty sender.
5142 39. The daemon was not analysing the content of -oX till after it had closed
5143 stderr and disconnected from the controlling terminal. This meant that any
5144 syntax errors were only noted on the panic log, and the return code from
5145 the command was 0. By re-arranging the code a little, I've made the
5146 decoding happen first, so such errors now appear on stderr, and the return
5147 code is 1. However, the actual setting up of the sockets still happens in
5148 the disconnected process, so errors there are still only recorded on the
5151 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5152 connections (as happens on some IP stacks) was logged at start up time as
5153 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5154 it from "IPv6 and IPv4", which means that two separate sockets are being
5157 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5158 says whether AF_INET or AF_INET6 was passed as an argument.
5160 42. Exiwhat output was messed up when time zones were included in log
5163 43. Exiwhat now gives more information about the daemon's listening ports,
5164 and whether -tls-on-connect was used.
5166 44. The "port" option of the smtp transport is now expanded.
5168 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5169 silently ignored. Now an error message is written to the main and panic
5172 46. There's a new ACL modifier called "logwrite" which writes to a log file
5173 as soon as it is encountered.
5175 47. Added $local_user_uid and $local_user_gid at routing time.
5177 48. Exim crashed when trying to verify a sender address that was being
5180 49. Exim was recognizing only a space character after ".include". It now also
5181 recognizes a tab character.
5183 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5184 extracting the relevant information from the specification. The man page no
5185 longer contains scrambled data for the -d option, and I've added a section
5186 at the front about calling Exim under different names.
5188 51. Added "extra_headers" argument to the "mail" command in filter files.
5190 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5193 53. Installed eximstats 1.29.
5195 54. Added transport_filter_timeout as a generic transport option.
5197 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5198 Cc: header lines. This was required by RFC 822, but it not required by RFC
5201 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5202 incoming messages that did not have them. Now it does so only if the
5203 message originates locally, that is, if there is no associated remote host
5204 address. When Resent- header lines are present, this applies to the Resent-
5205 lines rather than the non-Resent- lines.
5207 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5208 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5210 58. Messages for configuration errors now include the name of the main
5211 configuration file - useful now that there may be more than one file in a
5212 list (.included file names were always shown).
5214 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5215 for those rare installations that do not start the daemon as root or run it
5216 setuid root. I've cut out the call to initgroups() if the daemon is not
5219 60. The Exim user and group can now be bound into the binary as text strings
5220 that are looked up at the start of Exim's processing.
5222 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5224 62. Added $mailstore_basename variable.
5226 63. Installed patch to sieve.c from Michael Haardt.
5228 64. When Exim failed to open the panic log after failing to open the main log,
5229 the original message it was trying to log was written to stderr and debug
5230 output, but if they were not available (the usual case in production), it
5231 was lost. Now it is written to syslog before the two lines that record the
5232 failures to open the logs.
5234 65. Users' Exim filters run in subprocesses under the user's uid. It is
5235 possible for a "deliver" command or an alias in a "personal" command to
5236 provoke an address rewrite. If logging of address rewriting is configured,
5237 this fails because the process is not running as root or exim. There may be
5238 a better way of dealing with this, but for the moment (because 4.30 needs
5239 to be released), I have disabled address rewrite logging when running a
5240 filter in a non-root, non-exim process.
5246 1. The buildconfig auxiliary program wasn't quoting the value set for
5247 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5248 not defined. This bug was masked in 4.22 by the effect that was fixed in
5251 2. Some messages that were rejected after a message id was allocated were
5252 shown as "incomplete" by exigrep. It no longer does this for messages that
5253 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5255 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5256 have allow_domain_literals set, the ID did not get logged in the <= line.
5257 Domain literals are now always recognized in Message-ID: header lines.
5259 4. The first argument for a ${extract expansion item is the key name or field
5260 number. Leading and trailing spaces in this item were not being ignored,
5261 causing some misleading effects.
5263 5. When deliver_drop_privilege was set, single queue runner processes started
5264 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5265 same command in the process it spins off) were not dropping privilege.
5267 6. When the daemon running as "exim" started a queue runner, it always
5268 re-executed Exim in the spun-off process. This is a waste of effort when
5269 deliver_drop_privilege is set. The new process now just calls the
5270 queue-runner function directly.
5276 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5279 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5280 ignored. Though the use of -oP was forcing the writing of a pid file, it
5281 was always written to the default place.
5283 3. If the message "no IP address found for host xxxx" is generated during
5284 incoming verification, it is now followed by identification of the incoming
5285 connection (so you can more easily find what provoked it).
5287 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5289 5. Added some features to "harden" Exim a bit more against certain attacks:
5291 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5292 be put in Local/Makefile. This is like the never_users runtime option,
5293 but it cannot be overridden. The default setting is "root".
5295 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5296 prefix string with which any file named in a -C command line option
5299 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5300 is retained for -C and -D only if the caller of Exim is root. Without
5301 it, the exim user may also use -C and -D and retain privilege.
5303 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5304 command line option is disabled.
5306 6. Macro names set by the -D option must start with an upper case letter, just
5307 like macro names defined in the configuration file.
5309 7. Added "dereference=" facility to LDAP.
5311 8. Two instances of the typo "uknown" in the source files are fixed.
5313 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5314 the Configure-Makefile script screwed up while processing it.
5316 10. Incorporated PCRE 4.4.
5318 11. The SMTP synchronization check was not operating right at the start of an
5319 SMTP session. For example, it could not catch a HELO sent before the client
5320 waited for the greeting. There is now a check for outstanding input at the
5321 point when the greeting is written. Because of the duplex, asynchronous
5322 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5323 way, but not yet received, when the check is performed.
5325 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5326 on TCP/IP sockets, because this apparently causes some broken clients to
5329 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5330 unchanged) from the Cygwin maintainer.
5332 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5333 was supported instead of testing for mbx. Effectively a typo.
5335 15. The spa authenticator server code was not checking that the input it
5336 received was valid base64.
5338 16. The debug output line for the "set" modifier in ACLs was not showing the
5339 name of the variable that was being set.
5341 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5343 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5344 lookup on its own. Something else was needed to trigger the lookup. For
5345 example, a match in host_lookup or the need for a host name in a host list.
5346 Now, if $sender_host_name is referenced and the host name has not yet been
5347 looked up, a lookup is performed. If the lookup fails, the variable remains
5348 empty, and $host_lookup_failed is set to "1".
5350 19. Added "eqi" as a case-independent comparison operator.
5352 20. The saslauthd authentication condition could segfault if neither service
5353 nor realm was specified.
5355 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5356 error message that was logged was misleading, and incoming SMTP
5357 connections were dropped. The message is now more accurate, and temporary
5358 errors are given to SMTP connections.
5360 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5361 (except RSET) with a 421 error, until QUIT is received. However, it was
5362 failing to send a response to QUIT.
5364 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5365 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5366 ACL is now run after the helo_try_verify_hosts code.
5368 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5369 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5370 case-independent, so other case variants are also recognized). Apparently
5371 some systems use these upper case variants.
5373 25. If more than two messages were waiting for the same host, and a transport
5374 filter was specified for the transport, Exim sent two messages over the
5375 same TCP/IP connection, and then failed with "socket operation on non-
5376 socket" when it tried to send the third.
5378 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5380 27. The extern definition of crypt16() in expand.c was not being excluded when
5381 the OS had its own crypt16() function.
5383 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5384 as a preferred synonym for return_size_limit, both as an option and as an
5387 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5389 30. Changed the default configuration ACL to relax the local part checking rule
5390 for addresses that are not in any local domains. For these addresses,
5391 slashes and pipe symbols are allowed within local parts, but the sequence
5392 /../ is explicitly forbidden.
5394 31. SPA server authentication was not clearing the challenge buffer before
5397 32. log_message in a "warn" ACL statement was writing to the reject log as
5398 well as to the main log, which contradicts the documentation and doesn't
5399 seem right (because no rejection is happening). So I have stopped it.
5401 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5402 However, I am unable to do any testing of this.
5404 34. Fixed an infelicity in the appendfile transport. When checking directories
5405 for a mailbox, to see if any needed to be created, it was accidentally
5406 using path names with one or more superfluous leading slashes; tracing
5407 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5409 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5410 contents are added to the log line that is written for every discarded
5411 recipient. (Previously a log_message setting was ignored.)
5413 36. The ${quote: operator now quotes the string if it is empty.
5415 37. The install script runs exim in order to find its version number. If for
5416 some reason other than non-existence or emptiness, which it checks, it
5417 could not run './exim', it was installing it with an empty version number,
5418 i.e. as "exim-". This error state is now caught, and the installation is
5421 38. An argument was missing from the function that creates an error message
5422 when Exim fails to connect to the socket for saslauthd authentication.
5423 This could cause Exim to crash, or give a corrupted message.
5425 39. Added isip, isip4, and isip6 to ${if conditions.
5427 40. The ACL variables $acl_xx are now saved with the message, and can be
5428 accessed later in routers, transports, and filters.
5430 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
5431 strings in the file are not string-expanded.
5433 42. If a MAIL command specified a SIZE value that was too large to fit into an
5434 int variable, the check against message_size_limit failed. Such values are
5435 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
5436 day this will have to be increased, but I don't think I want to be around
5437 when emails are that large.
5444 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
5445 iconv() is not standard in FreeBSD.
5447 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
5448 IPv6 enabled. The observed symptom was a segmentation fault on return from
5449 the function os_common_find_running_interfaces() in src/os.c.
5451 3. In the check_special_case() function in daemon.c I had used "errno" as an
5452 argument name, which causes warnings on some systems. This was basically a
5453 typo, since it was named "eno" in the comments!
5455 4. The code that waits for the clock to tick (at a resolution of some fraction
5456 of a second) so as to ensure message-id uniqueness was always waiting for
5457 at least one whole tick, when it could have waited for less. [This is
5458 almost certainly not relevant at current processor speeds, where it is
5459 unlikely to ever wait at all. But we try to future-proof.]
5461 5. The function that sleeps for a time interval that includes fractions of a
5462 second contained a race. It did not block SIGALRM between setting the
5463 timer, and suspending (a couple of lines later). If the interval was short
5464 and the sigsuspend() was delayed until after it had expired, the suspension
5465 never ended. On busy systems this could lead to processes getting stuck for
5468 6. Some uncommon configurations may cause a lookup to happen in a queue runner
5469 process, before it forks any delivery processes. The open lookup caching
5470 mechanism meant that the open file or database connection was passed into
5471 the delivery process. The problem was that delivery processes always tidy
5472 up cached lookup data. This could cause a problem for the next delivery
5473 process started by the queue runner, because the external queue runner
5474 process does not know about the closure. So the next delivery process
5475 still has data in the lookup cache. In the case of a file lookup, there was
5476 no problem because closing a file descriptor in a subprocess doesn't affect
5477 the parent. However, if the lookup was caching a connection to a database,
5478 the connection was closed, and the second delivery process was likely to
5479 see errors such as "PGSQL: query failed: server closed the connection
5480 unexpectedly". The problem has been fixed by closing all cached lookups
5481 in a queue runner before running a delivery process.
5483 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
5484 seem to have the "const" qualifier which it has on other OS. I've
5487 8. Change 4.21/2 was too strict. It is only if there are two authenticators
5488 *of the same type* (client or server) with the same public name that an
5489 error should be diagnosed.
5491 9. When Exim looked up a host name for an IP address, but failed to find the
5492 original IP address when looking up the host name (a safety check), it
5493 output the message "<ip address> does not match any IP for NULL", which was
5494 confusing, to say the least. The bug was that the host name should have
5495 appeared instead of "NULL".
5497 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
5498 user that is built into the binary, and the -C or -D options is used, root
5499 privilege is dropped before the configuration file is read. In addition,
5500 logging is switched to stderr instead of the normal log files. If the
5501 configuration then re-defines the Exim user, the unprivileged environment
5502 is probably not what is expected, so Exim logs a panic warning message (but
5505 However, if deliver_drop_privilege is set, the unprivileged state may well
5506 be exactly what is intended, so the warning has been cut out in that case,
5507 and Exim is allowed to try to write to its normal log files.
5513 1. smtp_return_error_details was not giving details for temporary sender
5514 or receiver verification errors.
5516 2. Diagnose a configuration error if two authenticators have the same public
5519 3. Exim used not to create the message log file for a message until the first
5520 delivery attempt. This could be confusing when incoming messages were held
5521 for policy or load reasons. The message log file is now created at the time
5522 the message is received, and an initial "Received" line is written to it.
5524 4. The automatically generated man page for command line options had a minor
5525 bug that caused no ill effects; however, a more serious problem was that
5526 the procedure for building the man page automatically didn't always
5527 operate. Consequently, release 4.20 contains an out-of-date version. This
5528 shouldn't happen again.
5530 5. When building Exim with embedded Perl support, the script that builds the
5531 Makefile was calling 'perl' to find its compile-time parameters, ignoring
5532 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
5534 6. The freeze_tell option was not being used for messages that were frozen on
5535 arrival, either by an ACL or by local_scan().
5537 7. Added the smtp_incomplete_transaction log selector.
5539 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
5540 was accepting AUTH without a new EHLO.
5542 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
5543 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
5546 10. Logging of TCP/IP connections (when configured) now happens in the main
5547 daemon process instead of the child process, so that the TCP/IP connection
5548 count is more accurate (but it can never be perfect).
5550 11. The use of "drop" in a nested ACL was not being handled correctly in the
5551 outer ACL. Now, if condition failure induced by the nested "drop" causes
5552 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
5553 or "require"), the connection is dropped.
5555 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
5556 that yield "discard" can now be used with an "accept" or a "discard" verb,
5557 but an error is generated for any others (because I can't see a useful way
5558 to define what should happen).
5560 13. When an ACL is read dynamically from a file (or anywhere else), the lines
5561 are now processed in the same way as lines in the Exim configuration file.
5562 In particular, continuation lines are supported.
5564 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
5566 15. Added -ti meaning -t -i.
5568 16. Check for letters, digits, hyphens, and dots in the names of dnslist
5569 domains, and warn by logging if others are found.
5571 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
5572 returned from GIFCONF when Exim is trying to find the list of interfaces on
5573 a host. The code in os.c has been modified to copy each ifreq to an aligned
5574 structure in all cases.
5576 Also, in some cases, the returned ifreq's were being copied to a 'struct
5577 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
5578 means the last couple of bytes of an IPv6 address could be chopped if the
5579 ifreq contained only a normal sockaddr (14 bytes storage).
5581 18. Named domain lists were not supported in the hosts_treat_as_local option.
5582 An entry such as +xxxx was not recognized, and was treated as a literal
5585 19. Ensure that header lines added by a DATA ACL are included in the reject log
5586 if the ACL subsequently rejects the message.
5588 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
5589 MD5 (which is deprecated).
5591 21. When testing a filter file using -bf, Exim was writing a message when it
5592 took the sender from a "From " line in the message, but it was not doing so
5593 when it took $return_path from a Return-Path: header line. It now does.
5595 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
5596 with a valid header line field name (a series of printing characters
5597 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
5599 23. Changed "disc" in the source to "disk" to conform to the documentation and
5600 the book and for uniformity.
5602 24. Ignore Sendmail's -Ooption=value command line item.
5604 25. When execve() failed while trying to run a command in a pipe transport,
5605 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
5606 could be confused with a return value of 69 from the command itself. This
5607 has been changed to 127, the value the shell returns if it is asked to run
5608 a non-existent command. The wording for the related log line suggests a
5609 non-existent command as the problem.
5611 26. If received_header_text expands to an empty string, do not add a Received:
5612 header line to the message. (Well, it adds a token one on the spool, but
5613 marks it "old" so that it doesn't get used or transmitted.)
5615 27. Installed eximstats 1.28 (addition of -nt option).
5617 28. There was no check for failure on the call to getsockname() in the daemon
5618 code. This can fail if there is a shortage of resources on the system, with
5619 ENOMEM, for example. A temporary error is now given on failure.
5621 29. Contrary to the C standard, it seems that in some environments, the
5622 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
5623 program. Exim now does this explicitly; it affects the formatting of
5624 timestamps using strftime().
5626 30. If exiqsumm was given junk data, it threw up some uninitialized variable
5627 complaints. I've now initialized all the variables, to avoid this.
5629 32. Header lines added by a system filter were not being "seen" during
5630 transport-time rewrites.
5632 33. The info_callback() function passed to OpenSSL is set up with type void
5633 (*)(SSL *, int, int), as described somewhere. However, when calling the
5634 function (actually a macro) that sets it up, the type void(*)() is
5635 expected. I've put in a cast to prevent warnings from picky compilers.
5637 34. If a DNS black list lookup found a CNAME record, but there were no A
5638 records associated with the domain it pointed at, Exim crashed.
5640 35. If a DNS black list lookup returned more than one A record, Exim ignored
5641 all but the first. It now scans all returned addresses if a particular IP
5642 value is being sought. In this situation, the contents of the
5643 $dnslist_value variable are a list of all the addresses, separated by a
5646 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
5647 to accept a host name and all its aliases if the forward lookup for any of
5648 them yielded the IP address of the incoming connection. Now it accepts only
5649 those names whose forward lookup yields the correct IP address. Any other
5650 names are discarded. This closes a loophole whereby a rogue DNS
5651 administrator could create reverse DNS records to break through a
5652 wildcarded host restriction in an ACL.
5654 37. If a user filter or a system filter that ran in a subprocess used any of
5655 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
5656 the wrong values were passed to the pipe command ($thisaddress had the
5657 value of $0, $0 had the value of $1, etc). This bug was introduced by
5658 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
5660 38. Improved the line breaking for long SMTP error messages from ACLs.
5661 Previously, if there was no break point between 40 and 75 characters, Exim
5662 left the rest of the message alone. Two changes have been made: (a) I've
5663 reduced the minimum length to 35 characters; (b) if it can't find a break
5664 point between 35 and 75 characters, it looks ahead and uses the first one
5665 that it finds. This may give the occasional overlong line, but at least the
5666 remaining text gets split now.
5668 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
5669 file descriptors might be low, and that setting 1000 would always raise it.
5670 It turns out that in some environments, the limit is already over 1000 and
5671 that lowering it causes trouble. So now Exim takes care not to decrease it.
5673 40. When delivering a message, the value of $return_path is set to $sender_
5674 address at the start of routing (routers may change the value). By an
5675 oversight, this default was not being set up when an address was tested by
5676 -bt or -bv, which affected the outcome if any router or filter referred to
5679 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
5680 log only when "message" or "log_message" are set. However, if one of the
5681 conditions was an address verification, or a call to a nested ACL, the
5682 messages generated by the underlying test were being passed through. This
5683 no longer happens. The underlying message is available in $acl_verify_
5684 message for both "message" and "log_message" expansions, so it can be
5685 passed through if needed.
5687 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
5688 new expansion $bh_ to give the encoded byte string without charset
5689 translation. Translation happens only if iconv() is available; HAVE_ICONV
5690 indicates this at build time. HEADERS_CHARSET gives the charset to
5691 translate to; headers_charset can change it in the configuration, and
5692 "headers charset" can change it in an individual filter file.
5694 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
5695 that creates RFC 2047 encoded "words" labels them as that charset instead
5696 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
5697 expansion operator; (ii) when Exim creates a From: line for a local
5698 message; (iii) when a header line is rewritten to include a "phrase" part.
5700 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
5701 buggy, causing it to skip the first lines of messages whose message ID
5702 ended in 'D'. This would not have bitten before Exim release 4.14, because
5703 message IDs were unlikely to end in 'D' before then. The effect was to have
5704 incorrect size information for certain domains.
5706 45. #include "config.h" was missing at the start of the crypt16.c module. This
5707 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
5710 46. If there was a timeout during a "random" callout check, Exim treated it as
5711 a failure of the random address, and carried on sending RSET and the real
5712 address. If the delay was just some slowness somewhere, the response to the
5713 original RCPT would be taken as a response to RSET and so on, causing
5714 mayhem of various kinds.
5716 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
5717 when I implemented it. It didn't allow for the fact that some option values
5718 may legitimatetly be negative (e.g. size_addition), and it didn't even do
5719 the right test for positive values.
5721 48. Domain names in DNS records are case-independent. Exim always looks them up
5722 in lower case. Some resolvers return domain names in exactly the case they
5723 appear in the zone file, that is, they may contain uppercase letters. Not
5724 all resolvers do this - some return always lower case. Exim was treating a
5725 change of case by a resolver as a change of domain, similar to a widening
5726 of a domain abbreviation. This triggered its re-routing code and so it was
5727 trying to route what was effectively the same domain again. This normally
5728 caused routing to fail (because the router wouldn't handle the domain
5729 twice). Now Exim checks for this case specially, and just changes the
5730 casing of the domain that it ultimately uses when it transmits the message
5733 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
5736 50. If a filter generated a file delivery with a non-absolute name (possible if
5737 no home directory exists for the router), the forbid_file option was not
5740 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
5741 the existing equality matching.
5743 52. Exim was using ints instead of ino_t variables in some places where it was
5744 dealing with inode numbers.
5746 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
5747 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
5748 called TMPDIR, and if it finds it is different, it changes its value.
5750 54. The smtp_printf() function is now made available to local_scan() so
5751 additional output lines can be written before returning. There is also an
5752 smtp_fflush() function to enable the detection of a dropped connection.
5753 The variables smtp_input and smtp_batched_input are exported to
5756 55. Changed the default runtime configuration: the message "Unknown user"
5757 has been removed from the ACL, and instead placed on the localuser router,
5758 using the cannot_route_message feature. This means that any verification
5759 failures that generate their own messages won't get overridden. Similarly,
5760 the "Unrouteable address" message that was in the ACL for unverifiable
5761 relay addresses has also been removed.
5763 56. Added hosts_avoid_esmtp to the smtp transport.
5765 57. The exicyclog script was not checking for the esoteric option
5766 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
5767 will work only if exicyclog is run under the appropriate euid.
5769 58. Following a discussion on the list, the rules by which Exim recognises line
5770 endings on incoming messages have been changed. The -dropcr and drop_cr
5771 options are now no-ops, retained only for backwards compatibility. The
5772 following line terminators are recognized: LF CRLF CR. However, special
5773 processing applies to CR:
5775 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
5776 nor a local message in the state where . is a terminator.
5778 (ii) If a bare CR is encountered in a header line, an extra space is added
5779 after the line terminator so as not to end the header. The reasoning
5780 behind this is that bare CRs in header lines are most likely either
5781 to be mistakes, or people trying to play silly games.
5783 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
5784 was being incorrectly given as 18 bytes larger than it should have been.
5785 This is a VOB (very old bug).
5787 60. This may never have affected anything current, but just in case it has:
5788 When the local host is found other than at the start of a list of hosts,
5789 the local host, those with the same MX, and any that follow, are discarded.
5790 When the list in question was part of a longer list of hosts, the following
5791 hosts (not currently being processed) were also being discarded. This no
5792 longer happens. I'm not sure if this situation could ever has previously
5795 61. Added the "/MX" feature to lists of hosts in the manualroute and query
5798 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
5799 header. This is something that is recommended in a new Internet Draft, and
5800 is something that is documented as being done by Sendmail. There are two
5801 possible values. For messages generated by the autoreply transport, Exim
5804 Auto-Submitted: auto-replied
5806 whereas for all other generated messages (e.g. bounces) it adds
5808 Auto-Submitted: auto-generated
5810 63. The "personal" condition in filters now includes a test for the
5811 Auto-Submitted: header. If it contains the string "auto-" the message it
5812 not considered personal.
5814 64. Added rcpt_include_affixes as a generic transport option.
5816 65. Added queue_only_override (default true).
5818 66. Added the syslog_duplication option.
5820 67. If what should have been the first header line of a message consisted of
5821 a space followed by a colon, Exim was mis-interpreting it as a header line.
5822 It isn't of course - it is syntactically invalid and should therefore be
5823 treated as the start of the message body. The misbehaviour could have
5824 caused a number of strange effects, including loss of data in subsequent
5825 header lines, and spool format errors.
5827 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
5828 client host had authenticated. This control can now be exercised by an ACL
5829 for more flexibility.
5831 69. By default, callouts do not happen when testing with -bh. There is now a
5832 variant, -bhc, which does actually run the callout code, including
5833 consulting and updating the callout cache.
5835 70. Added support for saslauthd authentication, courtesy of Alexander
5838 71. If statvfs() failed on the spool or log directories while checking their
5839 size for availability, Exim confusingly gave the error "space shortage".
5840 Furthermore, in debugging mode it crashed with a floating point exception.
5841 These checks are done if check_{spool,log}_{space,inodes} are set, and when
5842 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
5843 serious problem, Exim now writes to the main and panic logs when this
5844 happens, with details of the failure. It then refuses to accept the
5845 incoming message, giving the message "spool directory problem" or "log
5846 directory problem" with a 421 code for SMTP messages.
5848 72. When Exim is about to re-exec itself, it ensures that the file descriptors
5849 0, 1, and 2 exist, because some OS complain for execs without them (see
5850 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
5851 descriptors. However, the code omitted to check that the open succeeded,
5852 causing mysterious errors if for some reason the permissions on /dev/null
5853 got screwed. Now Exim writes a message to the main and panic logs, and
5854 bombs out if it can't open /dev/null.
5856 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
5857 interact so that it is all more flexible. It is supposed to remain
5858 backwards compatible. Also added extra_local_interfaces.
5860 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
5861 to bomb out with an assertion failure - to the client this appears as a
5862 connection drop. This problem occurs in the part of the code that was taken
5863 from the Samba project. Fortunately, the assertion is in a very simple
5864 function, so I have fixed this by reproducing the function inline in the
5865 one place where it is called, and arranging for authentication to fail
5866 instead of killing the process with assert().
5868 75. The SPA client code was not working when the server requested OEM rather
5869 than Unicode encoding.
5871 76. Added code to make require_files with a specific uid setting more usable in
5872 the case where statting the file as root fails - usually a non-root-mounted
5873 NFS file system. When this happens and the failure is EACCES, Exim now
5874 forks a subprocess and does the per-uid checking as the relevant uid.
5876 77. Added process_log_path.
5878 78. If log_file_path was not explicitly set, a setting of check_log_space or
5879 check_log_inodes was ignored.
5881 79. If a space check for the spool or log partitions fails, the incident is now
5882 logged. Of course, in the latter case the data may get lost...
5884 80. Added the %p formatting code to string_format() so that it can be used to
5885 print addresses in debug_print(). Adjusted all the address printing in the
5886 debugging in store.c to use %p rather than %d.
5888 81. There was a concern that a line of code in smtp_in.c could overflow a
5889 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
5890 initially expressed, the concern was not well-founded, because trailing
5891 spaces are removed early. However, if the trailing spaces were followed by
5892 a NULL, they did not get removed, so the overflow was possible. Two fixes
5895 (a) I re-wrote the offending code in a cleaner fashion.
5896 (b) If an incoming SMTP command contains a NULL character, it is rejected
5899 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
5900 runs initgroups(), so that if the Exim user is in any additional groups,
5901 they will be used during message reception.
5907 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / heiko / exim.git / blob