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.
239 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
241 This, or similar/improved, will also be change PP/11 of 4.82.
247 PP/01 Handle short writes when writing local log-files.
248 In practice, only affects FreeBSD (8 onwards).
249 Bugzilla 1053, with thanks to Dmitry Isaikin.
251 NM/01 Bugzilla 949 - Documentation tweak
253 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
256 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
258 PP/02 Implemented gsasl authenticator.
260 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
262 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
263 `pkg-config foo` for cflags/libs.
265 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
266 with rest of GSASL and with heimdal_gssapi.
268 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
269 `pkg-config foo` for cflags/libs for the TLS implementation.
271 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
272 properties get this fed in as external SSF. A number of robustness
273 and debugging improvements to the cyrus_sasl authenticator.
275 PP/08 cyrus_sasl server now expands the server_realm option.
277 PP/09 Bugzilla 1214 - Log authentication information in reject log.
278 Patch by Jeremy Harris.
280 PP/10 Added dbmjz lookup type.
282 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
284 PP/12 MAIL args handles TAB as well as SP, for better interop with
285 non-compliant senders.
286 Analysis and variant patch by Todd Lyons.
288 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
289 Bug report from Lars Müller <lars@samba.org> (via SUSE),
290 Patch from Dirk Mueller <dmueller@suse.com>
292 PP/13 tls_peerdn now print-escaped for spool files.
293 Observed some $tls_peerdn in wild which contained \n, which resulted
294 in spool file corruption.
296 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
297 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
298 or write after TLS renegotiation, which otherwise led to messages
301 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
302 as a tracking header (ie: a signed header comes before the signature).
303 Patch from Wolfgang Breyha.
305 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
306 comma-sep list; embedded commas doubled.
308 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
310 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
312 Report and patch from Dmitry Banschikov.
314 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
315 Removed SSL_clear() after SSL_new() which led to protocol negotiation
316 failures. We appear to now support TLS1.1+ with Exim.
318 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
319 lets Exim select keys and certificates based upon TLS SNI from client.
320 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
321 before an outbound SMTP session. New log_selector, +tls_sni.
323 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
324 NULL dereference. Report and patch from Alun Jones.
326 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
327 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
328 Not seeing resolver debug output on NetBSD, but suspect this is a
329 resolver implementation change.
331 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
332 Left warnings. Added "eximon gdb" invocation mode.
334 PP/21 Defaulting "accept_8bitmime" to true, not false.
336 PP/22 Added -bw for inetd wait mode support.
338 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
339 locate the relevant includes and libraries. Made this the default.
341 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
342 Bugzilla 1246, report and most of solution from Tomasz Kusy.
344 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
345 This may cause build issues on older platforms.
347 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
348 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
349 gnutls_require_mac & gnutls_require_protocols (no longer supported).
350 Added SNI support via GnuTLS too.
351 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
353 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
355 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
356 Applied second patch from Janne, implementing suggestion to default
357 multiple-strings-in-record handling to match SPF spec.
359 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
361 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
362 read-only, out of scope).
363 Patch from Wolfgang Breyha, report from Stuart Northfield.
365 PP/29 Fix three issues highlighted by clang analyser static analysis.
366 Only crash-plausible issue would require the Cambridge-specific
367 iplookup router and a misconfiguration.
368 Report from Marcin Mirosław.
370 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
372 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
373 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
374 As part of this, removing so much warning spew let me fix some minor
375 real issues in debug logging.
377 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
378 assignment on my part. Fixed.
380 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
381 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
382 Janne Snabb (who went above and beyond: thank you).
384 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
385 string otherwise requires a connection and a bunch more work and it's
386 relatively easy to get wrong. Should also expose TLS library linkage
389 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
390 64-bit ${eval} (JH/03).
392 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
393 GNU libc to support some of the 64-bit stuff, should not lead to
394 conflicts. Defined before os.h is pulled in, so if a given platform
395 needs to override this, it can.
397 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
398 protection layer was required, which is not implemented.
399 Bugzilla 1254, patch from Wolfgang Breyha.
401 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
402 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
403 tls_dhparam take prime identifiers. Also unbreak combination of
404 OpenSSL+DH_params+TLSSNI.
406 PP/39 Disable SSLv2 by default in OpenSSL support.
412 PP/01 Solaris build fix for Oracle's LDAP libraries.
413 Bugzilla 1109, patch from Stephen Usher.
415 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
417 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
420 TF/02 Fix a couple more cases where we did not log the error message
421 when unlink() failed. See also change 4.74-TF/03.
423 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
424 lock up or crash if it happened to be inside a call to libc when it
425 got a SIGUSR1 from exiwhat.
427 The SIGUSR1 handler appends the current process status to the process
428 log which is later printed by exiwhat. It used to use the general
429 purpose logging code to do this, but several functions it calls are
430 not safe for signals.
432 The new output code in the SIGUSR1 handler is specific to the process
433 log, and simple enough that it's easy to inspect for signal safety.
434 Removing some special cases also simplifies the general logging code.
435 Removing the spurious timestamps from the process log simplifies
438 TF/04 Improved ratelimit ACL condition.
440 The /noupdate option has been deprecated in favour of /readonly which
441 has clearer semantics. The /leaky, /strict, and /readonly update modes
442 are mutually exclusive. The update mode is no longer included in the
443 database key; it just determines when the database is updated. (This
444 means that when you upgrde Exim will forget old rate measurements.)
446 Exim now checks that the per_* options are used with an update mode that
447 makes sense for the current ACL. For example, when Exim is processing a
448 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
449 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
450 must specify per_mail/readonly. If you omit the update mode it defaults to
451 /leaky where that makes sense (as before) or /readonly where required.
453 The /noupdate option is now undocumented but still supported for
454 backwards compatibility. It is equivalent to /readonly except that in
455 ACLs where /readonly is required you may specify /leaky/noupdate or
456 /strict/noupdate which are treated the same as /readonly.
458 A useful new feature is the /count= option. This is a generalization
459 of the per_byte option, so that you can measure the throughput of other
460 aggregate values. For example, the per_byte option is now equivalent
461 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
463 The per_rcpt option has been generalized using the /count= mechanism
464 (though it's more complicated than the per_byte equivalence). When it is
465 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
466 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
467 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
468 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
469 Note that using per_rcpt with a non-readonly update mode in more than
470 one ACL will cause the recipients to be double-counted. (The per_mail
471 and per_byte options don't have this problem.)
473 The handling of very low rates has changed slightly. If the computed rate
474 is less than the event's count (usually one) then this event is the first
475 after a long gap. In this case the rate is set to the same as this event's
476 count, so that the first message of a spam run is counted properly.
478 The major new feature is a mechanism for counting the rate of unique
479 events. The new per_addr option counts the number of different
480 recipients that someone has sent messages to in the last time period. It
481 behaves like per_rcpt if all the recipient addresses are different, but
482 duplicate recipient addresses do not increase the measured rate. Like
483 the /count= option this is a general mechanism, so the per_addr option
484 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
485 example, measure the rate that a client uses different sender addresses
486 with the options per_mail/unique=$sender_address. There are further
487 details in the main documentation.
489 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
491 TF/06 Removed a few PCRE remnants.
493 TF/07 Automatically extract Exim's version number from tags in the git
494 repository when doing development or release builds.
496 PP/02 Raise smtp_cmd_buffer_size to 16kB.
497 Bugzilla 879. Patch from Paul Fisher.
499 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
500 Heavily based on revision 40f9a89a from Simon Arlott's tree.
503 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
505 PP/05 Variable $av_failed, true if the AV scanner deferred.
506 Bugzilla 1078. Patch from John Horne.
508 PP/06 Stop make process more reliably on build failure.
509 Bugzilla 1087. Patch from Heiko Schlittermann.
511 PP/07 Make maildir_use_size_file an _expandable_ boolean.
512 Bugzilla 1089. Patch from Heiko Schlittermann.
514 PP/08 Handle ${run} returning more data than OS pipe buffer size.
515 Bugzilla 1131. Patch from Holger Weiß.
517 PP/09 Handle IPv6 addresses with SPF.
518 Bugzilla 860. Patch from Wolfgang Breyha.
520 PP/10 GnuTLS: support TLS 1.2 & 1.1.
522 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
525 PP/11 match_* no longer expand right-hand-side by default.
526 New compile-time build option, EXPAND_LISTMATCH_RHS.
527 New expansion conditions, "inlist", "inlisti".
529 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
531 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
533 PP/14 fix log_write() format string regression from TF/03.
534 Bugzilla 1152. Patch from Dmitry Isaikin.
540 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
542 PP/02 Harmonised TLS library version reporting; only show if debugging.
543 Layout now matches that introduced for other libraries in 4.74 PP/03.
545 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
547 PP/04 New "dns_use_edns0" global option.
549 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
552 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
553 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
555 TK/01 Updated PolarSSL code to 0.14.2.
556 Bugzilla 1097. Patch from Andreas Metzler.
558 PP/07 Catch divide-by-zero in ${eval:...}.
561 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
564 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
565 format-string attack -- SECURITY: remote arbitrary code execution.
567 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
568 time unintentionally subject to list matching rules, letting the header
569 cause arbitrary Exim lookups (of items which can occur in lists, *not*
570 arbitrary string expansion). This allowed for information disclosure.
572 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
573 INT_MIN/-1 -- value coerced to INT_MAX.
579 NM/01 Workround for PCRE version dependency in version reporting
582 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
583 This fixes portability to compilers other than gcc, notably
584 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
586 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
587 makefiles for portability to HP-UX and POSIX correctness.
589 PP/01 Permit LOOKUP_foo enabling on the make command-line.
590 Also via indented variable definition in the Makefile.
591 (Debugging by Oliver Heesakkers).
593 PP/02 Restore caching of spamd results with expanded spamd_address.
594 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
596 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
597 Improves build reliability. Fix from: Frank Elsner
599 NM/02 Fix wide character breakage in the rfc2047 coding
600 Fixes bug 1064. Patch from Andrey N. Oktyabrski
602 NM/03 Allow underscore in dnslist lookups
603 Fixes bug 1026. Patch from Graeme Fowler
605 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
606 Code patches from Adam Ciarcinski of NetBSD.
608 NM/04 Fixed exiqgrep to cope with mailq missing size issue
611 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
612 is logged, to avoid truncation. Patch from John Horne.
614 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
615 Patch from Jakob Hirsch.
617 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
618 SQL string expansion failure details.
619 Patch from Andrey Oktyabrski.
621 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
622 Patch from Simon Arlott.
624 PP/09 New lookups functionality failed to compile on old gcc which rejects
625 extern declarations in function scope.
626 Patch from Oliver Fleischmann
628 PP/10 Use sig_atomic_t for flags set from signal handlers.
629 Check getgroups() return and improve debugging.
630 Fixed developed for diagnosis in bug 927 (which turned out to be
633 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
634 Patch from Mark Zealey.
636 PP/12 Bugzilla 1056: Improved spamd server selection.
637 Patch from Mark Zealey.
639 PP/13 Bugzilla 1086: Deal with maildir quota file races.
640 Based on patch from Heiko Schlittermann.
642 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
643 Patch from Uwe Doering, sign-off by Michael Haardt.
645 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
646 variable declaration deep within a block. Bug and patch from
649 PP/15 lookups-Makefile IRIX compatibilty coercion.
651 PP/16 Make DISABLE_DKIM build knob functional.
653 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
654 Patch by Simon Arlott
656 TF/03 Fix valgrind.h portability to C89 compilers that do not support
657 variable argument macros. Our copy now differs from upstream.
663 TF/01 Failure to get a lock on a hints database can have serious
664 consequences so log it to the panic log.
666 TF/02 Log LMTP confirmation messages in the same way as SMTP,
667 controlled using the smtp_confirmation log selector.
669 TF/03 Include the error message when we fail to unlink a spool file.
671 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
672 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
673 for maintaining out-of-tree patches for some time.
675 PP/01 Bugzilla 139: Documentation and portability issues.
676 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
677 Handle per-OS dynamic-module compilation flags.
679 PP/02 Let /dev/null have normal permissions.
680 The 4.73 fixes were a little too stringent and complained about the
681 permissions on /dev/null. Exempt it from some checks.
682 Reported by Andreas M. Kirchwitz.
684 PP/03 Report version information for many libraries, including
685 Exim version information for dynamically loaded libraries. Created
686 version.h, now support a version extension string for distributors
687 who patch heavily. Dynamic module ABI change.
689 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
690 privilege escalation vulnerability whereby the Exim run-time user
691 can cause root to append content of the attacker's choosing to
694 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
697 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
698 If dropping privileges for untrusted macros, we disabled normal logging
699 on the basis that it would fail; for the Exim run-time user, this is not
700 the case, and it resulted in successful deliveries going unlogged.
701 Fixed. Reported by Andreas Metzler.
707 PP/01 Date: & Message-Id: revert to normally being appended to a message,
708 only prepend for the Resent-* case. Fixes regression introduced in
709 Exim 4.70 by NM/22 for Bugzilla 607.
711 PP/02 Include check_rfc2047_length in configure.default because we're seeing
712 increasing numbers of administrators be bitten by this.
714 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
716 PP/03 Bugzilla 994: added openssl_options main configuration option.
718 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
720 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
722 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
724 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
725 without a peer certificate, leading to a segfault because of an
726 assumption that peers always have certificates. Be a little more
727 paranoid. Problem reported by Martin Tscholak.
729 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
730 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
731 NB: ClamAV planning to remove STREAM in "middle of 2010".
732 CL also introduces -bmalware, various -d+acl logging additions and
733 more caution in buffer sizes.
735 PP/09 Implemented reverse_ip expansion operator.
737 PP/10 Bugzilla 937: provide a "debug" ACL control.
739 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
741 PP/12 Bugzilla 973: Implement --version.
743 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
745 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
747 PP/15 Bugzilla 816: support multiple condition rules on Routers.
749 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
750 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
751 ignore trailing whitespace.
753 JJ/02 prevent non-panic DKIM error from being sent to paniclog
755 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
758 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
759 Notification from Dr Andrew Aitchison.
761 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
762 ExtendedDetectionInfo response format.
763 Notification from John Horne.
765 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
768 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
769 XSL and documented dependency on system catalogs, with examples of how
772 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
775 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
776 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
777 configuration file which is writeable by the Exim user or group.
779 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
780 of configuration files to cover files specified with the -C option if
781 they are going to be used with root privileges, not just the default
784 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
785 option (effectively making it always true).
787 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
788 files to be used while preserving root privileges.
790 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
791 that rogue child processes cannot use them.
793 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
794 run-time user, instead of root.
796 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
797 Exim run-time user without dropping privileges.
799 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
800 result string, instead of calling string_vformat() twice with the same
803 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
804 for other users. Others should always drop root privileges if they use
805 -C on the command line, even for a whitelisted configure file.
807 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
809 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
815 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
816 $data_path, and $header_path variables; fixed documentation bugs and
819 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
820 exipick to access non-standard spools, including the "frozen" queue
823 NM/01 Bugzilla 965: Support mysql stored procedures.
824 Patch from Alain Williams
826 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
828 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
829 Patch from Andreas Metzler
831 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
832 Patch from Kirill Miazine
834 NM/05 Bugzilla 671: Added umask to procmail example.
836 JJ/03 installed exipick 20100323.0, fixing doc bug
838 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
839 directory. Notification and patch from Dan Rosenberg.
841 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
843 TK/02 Improve log output when DKIM signing operation fails.
845 MH/01 Treat the transport option dkim_domain as a colon separated
846 list, not as a single string, and sign the message with each element,
847 omitting multiple occurences of the same signer.
849 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
850 Bugzilla 985, 986. Patch by Simon Arlott
852 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
853 Patch by Simon Arlott
855 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
856 MBX locking. Notification from Dan Rosenberg.
862 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
864 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
866 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
868 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
870 NM/04 Bugzilla 915: Flag broken perl installation during build.
876 TK/01 Added patch by Johannes Berg that expands the main option
877 "spamd_address" if it starts with a dollar sign.
879 TK/02 Write list of recipients to X-Envelope-Sender header when building
880 the mbox-format spool file for content scanning (suggested by Jakob
883 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
884 (http://www.dcc-servers.net/) support via dccifd. Activated by
885 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
887 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
888 by Mark Daniel Reidel <mr@df.eu>.
890 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
891 When building exim an external PCRE library is now needed -
892 PCRE is a system library on the majority of modern systems.
893 See entry on PCRE_LIBS in EDITME file.
895 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
896 conversation. Added nologin parameter to request.
897 Patch contributed by Kirill Miazine.
899 TF/01 Do not log submission mode rewrites if they do not change the address.
901 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
903 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
904 log files in place. Contributed by Roberto Lima.
906 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
908 TF/03 Bugzilla 615: When checking the local_parts router precondition
909 after a local_part_suffix or local_part_prefix option, Exim now
910 does not use the address's named list lookup cache, since this
911 contains cached lookups for the whole local part.
913 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
914 Robert Millan. Documentation is in experimental-spec.txt.
916 TF/04 Bugzilla 668: Fix parallel build (make -j).
918 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
920 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
921 Patch provided by Jan Srzednicki.
923 TF/05 Leading white space used to be stripped from $spam_report which
924 wrecked the formatting. Now it is preserved.
926 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
927 that they are available at delivery time.
929 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
931 TF/08 TLS error reporting now respects the incoming_interface and
932 incoming_port log selectors.
934 TF/09 Produce a more useful error message if an SMTP transport's hosts
935 setting expands to an empty string.
937 NM/06 Bugzilla 744: EXPN did not work under TLS.
938 Patch provided by Phil Pennock.
940 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
941 Patch provided by Richard Godbee.
943 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
944 acl_smtp_notquit, added index entry.
946 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
947 Patch provided by Eugene Bujak.
949 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
950 accept(). Patch provided by Maxim Dounin.
952 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
953 Patch provided by Phil Pennock.
955 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
957 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
958 Patch provided by Brad "anomie" Jorsch.
960 NM/14 Bugzilla 622: Added timeout setting to transport filter.
961 Patch provided by Dean Brooks.
963 TK/05 Add native DKIM support (does not depend on external libraries).
965 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
966 Patch provided by Graeme Fowler.
968 NM/16 Bugzilla 851: Documentation example syntax fix.
970 NM/17 Changed NOTICE file to remove references to embedded PCRE.
972 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
975 NM/19 Bugzilla 745: TLS version reporting.
976 Patch provided by Phil Pennock.
978 NM/20 Bugzilla 167: bool: condition support.
979 Patch provided by Phil Pennock.
981 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
982 clients. Patch provided by Phil Pennock.
984 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
985 Patch provided by Brad "anomie" Jorsch.
987 NM/23 Bugzilla 687: Fix misparses in eximstats.
988 Patch provided by Heiko Schlittermann.
990 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
991 Patch provided by Heiko Schlittermann.
993 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
994 plus update to original patch.
996 NM/26 Bugzilla 799: Documentation correction for ratelimit.
998 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
999 Patch provided by David Brownlee.
1001 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1003 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1005 NM/30 Bugzilla 888: TLS documentation bugfixes.
1007 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1009 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1010 Unlike the original bugzilla I have changed all shell scripts in src tree.
1012 NM/33 Bugzilla 898: Transport filter timeout fix.
1013 Patch by Todd Rinaldo.
1015 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1016 Patch by Serge Demonchaux.
1018 NM/35 Bugzilla 39: Base64 decode bug fixes.
1019 Patch by Jakob Hirsch.
1021 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1023 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1025 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1027 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1033 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1034 ALT-N's libdkim that I have put here:
1035 http://duncanthrax.net/exim-experimental/
1037 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1038 were called 'true' and it seems that C99 defines that as a reserved
1039 keyword to be used with 'bool' variable types. That means you could
1040 not include C99-style headers which use bools without triggering
1041 build errors in sieve.c.
1043 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1044 as mailq or other aliases. Changed the --help handling significantly
1045 to do whats expected. exim_usage() emits usage/help information.
1047 SC/01 Added the -bylocaldomain option to eximstats.
1049 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1051 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1053 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1059 PH/01 Another patch from the Sieve maintainer.
1061 PH/02 When an IPv6 address is converted to a string for single-key lookup
1062 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1063 dots are used instead of colons so that keys in lsearch files need not
1064 contain colons. This was done some time before quoting was made available
1065 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1066 (notated using the quote facility) so as to distinguish them from IPv4
1067 keys. This meant that lookups for IP addresses in host lists did not work
1068 for iplsearch lookups.
1070 This has been fixed by arranging for IPv6 addresses to be expressed with
1071 colons if the lookup type is iplsearch. This is not incompatible, because
1072 previously such lookups could never work.
1074 The situation is now rather anomolous, since one *can* have colons in
1075 ordinary lsearch keys. However, making the change in all cases is
1076 incompatible and would probably break a number of configurations.
1078 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1081 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1082 conversion specification without a maximum field width, thereby enabling
1083 a rogue spamd server to cause a buffer overflow. While nobody in their
1084 right mind would setup Exim to query an untrusted spamd server, an
1085 attacker that gains access to a server running spamd could potentially
1086 exploit this vulnerability to run arbitrary code as the Exim user.
1088 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1089 $primary_hostname instead of what libspf2 thinks the hosts name is.
1091 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1092 a directory entry by the name of the lookup key. Previously, if a
1093 symlink pointed to a non-existing file or a file in a directory that
1094 Exim lacked permissions to read, a lookup for a key matching that
1095 symlink would fail. Now it is enough that a matching directory entry
1096 exists, symlink or not. (Bugzilla 503.)
1098 PH/03 The body_linecount and body_zerocount variables are now exported in the
1101 PH/04 Added the $dnslist_matched variable.
1103 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1104 This means they are set thereafter only if the connection becomes
1107 PH/06 Added the client_condition to authenticators so that some can be skipped
1108 by clients under certain conditions.
1110 PH/07 The error message for a badly-placed control=no_multiline_responses left
1111 "_responses" off the end of the name.
1113 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1115 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1116 (without spaces) instead of just copying the configuration text.
1118 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1120 PH/11 Added $max_received_linelength.
1122 PH/12 Added +ignore_defer and +include_defer to host lists.
1124 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1125 way in which PCRE > 7.0 is built.
1127 PH/14 Implemented queue_only_load_latch.
1129 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1130 MAIL command. The effect was to mangle the value on 64-bit systems.
1132 PH/16 Another patch from the Sieve maintainer.
1134 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1136 PH/18 If a system quota error occurred while trying to create the file for
1137 a maildir delivery, the message "Mailbox is full" was not appended to the
1138 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1139 only to a quota excession during the actual writing of the file.
1141 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1142 characters?) which causes problems in log lines. The DN values are now
1143 passed through string_printing() before being added to log lines.
1145 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1146 and InterBase are left for another time.)
1148 PH/21 Added message_body_newlines option.
1150 PH/22 Guard against possible overflow in moan_check_errorcopy().
1152 PH/23 POSIX allows open() to be a macro; guard against that.
1154 PH/24 If the recipient of an error message contained an @ in the local part
1155 (suitably quoted, of course), incorrect values were put in $domain and
1156 $local_part during the evaluation of errors_copy.
1162 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1163 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1166 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1167 issue a MAIL command.
1169 PH/02 In an ACL statement such as
1171 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1173 if a client was not listed at all, or was listed with a value other than
1174 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1175 the condition was not true (as it should be), so access was not denied.
1176 The bug was that the ! inversion was incorrectly passed on to the second
1177 item. This has been fixed.
1179 PH/03 Added additional dnslists conditions == and =& which are different from
1180 = and & when the dns lookup returns more than one IP address.
1182 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1183 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1185 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1186 FSYNC, which compiles an option called disable_fsync that allows for
1187 bypassing fsync(). The documentation is heavily laced with warnings.
1189 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1191 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1192 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1193 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1194 including adding "make clean"; (3) Added -fPIC when compiling the test
1195 dynamically loaded module, to get rid of a warning.
1197 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1198 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1199 = 0s. The bug is otherwise harmless.
1201 PH/07 There was a bug in the dovecot authenticator such that the value of
1202 $auth1 could be overwritten, and so not correctly preserved, after a
1203 successful authentication. This usually meant that the value preserved by
1204 the server_setid option was incorrect.
1206 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1208 PH/09 Installed PCRE release 7.0.
1210 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1211 run for batched SMTP input. It is now run at the start of every message
1212 in the batch. While fixing this I discovered that the process information
1213 (output by running exiwhat) was not always getting set for -bs and -bS
1214 input. This is fixed, and it now also says "batched" for BSMTP.
1216 PH/11 Added control=no_pipelining.
1218 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1219 patch, slightly modified), and move the expansion of helo_data till after
1220 the connection is made in the smtp transport (so it can use these
1223 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1225 PH/14 Added log_selector = +pid.
1227 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1229 PH/16 Add ${if forany and ${if forall.
1231 PH/17 Added dsn_from option to vary the From: line in DSNs.
1233 PH/18 Flush SMTP output before performing a callout, unless control =
1234 no_callout_flush is set.
1236 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1237 was true (the default) a successful delivery failed to delete the retry
1238 item, thus causing premature timeout of the address. The bug is now
1241 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1243 PH/21 Long custom messages for fakedefer and fakereject are now split up
1244 into multiline reponses in the same way that messages for "deny" and
1245 other ACL rejections are.
1247 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1248 with slight modification.
1250 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1251 draft, changing the syntax and factoring some duplicate code".
1253 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1254 for deliveries of the second and subsequent messages over the same SMTP
1257 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1258 ${reduce, with only minor "tidies".
1260 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1262 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1263 expansion side effects.
1265 PH/27 When a message times out after an over-quota error from an Exim-imposed
1266 quota, the bounce message says "mailbox is full". This message was not
1267 being given when it was a system quota that was exceeded. It now should
1270 MH/03 Made $recipients available in local_scan(). local_scan() already has
1271 better access to the recipient list through recipients_list[], but
1272 $recipients can be useful in postmaster-provided expansion strings.
1274 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1275 in the case of a MAIL command with additional options following the
1276 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1277 were accidentally chopped off.
1279 PH/29 SMTP synchronization checks are implemented when a command is read -
1280 there is a check that no more input is waiting when there shouldn't be
1281 any. However, for some commands, a delay in an ACL can mean that it is
1282 some time before the response is written. In this time, more input might
1283 arrive, invalidly. So now there are extra checks after an ACL has run for
1284 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1285 pipelining has not been advertised.
1287 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1289 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1290 returned in the SMTP response when smtp_return_error_details was false.
1291 This has been fixed.
1293 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1294 instead of the C I/O that was originally supplied, because problems were
1295 reported on Solaris.
1297 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1298 Exim which did not show up earlier: it was assuming that a call to
1299 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1300 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1301 was a macro that became an assignment, so it seemed to work. This has
1302 changed to a proper function call with a void return, hence the compile
1303 error. Exim's code has been fixed.
1305 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1308 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1310 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1312 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1313 because it turns out to be too big or there is a timeout) there is no
1314 "Completed" line in the log. When some messages of this type were
1315 selected by exigrep, they were listed as "not completed". Others were
1316 picked up by some special patterns. I have improved the selection
1317 criteria to be more general.
1319 PH/38 The host_find_failed option in the manualroute router can now be set
1320 to "ignore", to completely ignore a host whose IP address cannot be
1321 found. If all hosts are ignored, the behaviour is controlled by the new
1322 host_all_ignored option.
1324 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1325 homing or because of multiple MX records with /mx) generated more than
1326 one IP address, and the following item turned out to be the local host,
1327 all the secondary addresses of the first item were incorrectly removed
1328 from the list, along with the local host and any following hosts (which
1329 is what is supposed to happen).
1331 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1332 whoever called Exim into the -H file. In the case of the daemon it was
1333 behaving confusingly. When first started, it used values for whoever
1334 started the daemon, but after a SIGHUP it used the Exim user (because it
1335 calls itself on a restart). I have changed the code so that it now always
1338 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1339 message are rejected with the same error (e.g. no authentication or bad
1340 sender address), and a DATA command is nevertheless sent (as can happen
1341 with PIPELINING or a stupid MUA), the error message that was given to the
1342 RCPT commands is included in the rejection of the DATA command. This is
1343 intended to be helpful for MUAs that show only the final error to their
1346 PH/42 Another patch from the Sieve maintainer.
1348 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1349 Eximstats - Fixed some broken HTML links and added missing column headers
1351 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1352 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1354 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1356 PH/43 Yet another patch from the Sieve maintainer.
1358 PH/44 I found a way to check for a TCP/IP connection going away before sending
1359 the response to the final '.' that terminates a message, but only in the
1360 case where the client has not sent further data following the '.'
1361 (unfortunately, this is allowed). However, in many cases there won't be
1362 any further data because there won't be any more messages to send. A call
1363 to select() can be used: if it shows that the input is "ready", there is
1364 either input waiting, or the socket has been closed. An attempt to read
1365 the next input character can distinguish the two cases. Previously, Exim
1366 would have sent an OK response which the client would never have see.
1367 This could lead to message repetition. This fix should cure that, at
1368 least in a lot of common cases.
1370 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1371 advertised in response to EHLO.
1377 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1378 fixed by 4.65/MH/01 (is this a record?) are fixed:
1380 (i) An empty string was always treated as zero by the numeric comparison
1381 operators. This behaviour has been restored.
1383 (ii) It is documented that the numeric comparison operators always treat
1384 their arguments as decimal numbers. This was broken in that numbers
1385 starting with 0 were being interpreted as octal.
1387 While fixing these problems I realized that there was another issue that
1388 hadn't been noticed. Values of message_size_limit (both the global option
1389 and the transport option) were treated as octal if they started with 0.
1390 The documentation was vague. These values are now always treated as
1391 decimal, and I will make that clear in the documentation.
1397 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1398 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1401 MH/01 Don't check that the operands of numeric comparison operators are
1402 integers when their expansion is in "skipping" mode (fixes bug
1403 introduced by 4.64-PH/07).
1405 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1406 child addresses, Exim now panics and dies. Previously, because the count
1407 is held in a short int, deliveries were likely to be lost. As such a
1408 large number of recipients for a single message is ridiculous
1409 (performance will be very, very poor), I have chosen to impose a limit
1410 rather than extend the field.
1416 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1417 leftover -K file (the existence of which was triggered by #402).
1418 While we were at it, introduced process PID as part of the -K
1419 filename. This should rule out race conditions when creating
1422 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1423 processing considerably. Previous code took too long for large mails,
1424 triggering a timeout which in turn triggers #401.
1426 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1427 in the DK code in transports.c. sendfile() is not really portable,
1428 hence the _LINUX specificness.
1430 TF/01 In the add_headers option to the mail command in an Exim filter,
1431 there was a bug that Exim would claim a syntax error in any
1432 header after the first one which had an odd number of characters
1435 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1436 callout verification, Exim cached a "reject" for the entire domain. This
1437 is correct for most verifications, but it is not correct for a recipient
1438 verification with use_sender or use_postmaster set, because in that case
1439 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1440 case of MAIL FROM:<> rejection from other early rejections (e.g.
1441 rejection of HELO). When verifying a recipient using a non-null MAIL
1442 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1443 Whatever the result of the callout, the value of the domain cache is
1444 left unchanged (for any other kind of callout, getting as far as trying
1445 RCPT means that the domain itself is ok).
1447 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
1450 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
1451 manifest itself as EPIPE rather than ECONNECT. When tidying away a
1452 session, the daemon ignores ECONNECT errors and logs others; it now
1453 ignores EPIPE as well.
1455 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
1456 (quoted-printable decoding).
1458 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
1459 later the small subsequent patch to fix an introduced bug.
1461 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
1463 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
1465 PH/08 An error is now given if message_size_limit is specified negative.
1467 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
1468 to be given (somewhat) arbitrary names.
1470 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
1473 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
1474 miscellaneous code fixes
1476 PH/10 Added the log_reject_target ACL modifier to specify where to log
1479 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
1480 hostname. This is wrong, because it relates to the incoming message (and
1481 probably the interface on which it is arriving) and not to the outgoing
1482 callout (which could be using a different interface). This has been
1483 changed to use the value of the helo_data option from the smtp transport
1484 instead - this is what is used when a message is actually being sent. If
1485 there is no remote transport (possible with a router that sets up host
1486 addresses), $smtp_active_hostname is used.
1488 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
1489 tweaks were necessary in order to get it to work (see also 21 below):
1490 (a) The code assumed that strncpy() returns a negative number on buffer
1491 overflow, which isn't the case. Replaced with Exim's string_format()
1493 (b) There were several signed/unsigned issues. I just did the minimum
1494 hacking in of casts. There is scope for a larger refactoring.
1495 (c) The code used strcasecmp() which is not a standard C function.
1496 Replaced with Exim's strcmpic() function.
1497 (d) The code set only $1; it now sets $auth1 as well.
1498 (e) A simple test gave the error "authentication client didn't specify
1499 service in request". It would seem that Dovecot has changed its
1500 interface. Fortunately there's a specification; I followed it and
1501 changed what the client sends and it appears to be working now.
1503 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
1506 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
1507 address A is aliased to B and C, where B exists and C does not. Without
1508 -v the output is "A verified" because verification stops after a
1509 successful redirection if more than one address is generated. However,
1510 with -v the child addresses are also verified. Exim was outputting "A
1511 failed to verify" and then showing the successful verification for C,
1512 with its parentage. It now outputs "B failed to verify", showing B's
1513 parentage before showing the successful verification of C.
1515 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
1516 look up a TXT record in a specific list after matching in a combined
1519 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
1520 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
1521 they consult the DNS. I had assumed they would set it the way they
1522 wanted; and indeed my experiments on Linux seem to show that in some
1523 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
1524 To be on the safe side, however, I have now made the interface to
1525 host_find_byname() similar to host_find_bydns(), with an argument
1526 containing the DNS resolver options. The host_find_byname() function now
1527 sets these options at its start, just as host_find_bydns() does. The smtp
1528 transport options dns_qualify_single and dns_search_parents are passed to
1529 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
1530 of host_find_byname() use the default settings of RES_DEFNAMES
1531 (qualify_single) but not RES_DNSRCH (search_parents).
1533 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
1534 spool_read_header() do less string testing, by means of a preliminary
1535 switch on the second character of optional "-foo" lines. (This is
1536 overdue, caused by the large number of possibilities that now exist.
1537 Originally there were few.) While I was there, I also converted the
1538 str(n)cmp tests so they don't re-test the leading "-" and the first
1539 character, in the hope this might squeeze out yet more improvement.
1541 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
1542 flag allowing group syntax was set by the header_syntax check but not
1543 turned off, possible causing trouble later; (2) The flag was not being
1544 set at all for the header_verify test, causing "group"-style headers to
1545 be rejected. I have now set it in this case, and also caused header_
1546 verify to ignore an empty address taken from a group. While doing this, I
1547 came across some other cases where the code for allowing group syntax
1548 while scanning a header line wasn't quite right (mostly, not resetting
1549 the flag correctly in the right place). These bugs could have caused
1550 trouble for malformed header lines. I hope it is now all correct.
1552 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
1553 with the "reply" argument non-NULL. The code, however (which originally
1554 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
1555 but it didn't always do it. This confused somebody who was copying the
1556 code for some other use. I have removed all the tests.
1558 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
1559 feature that was used to support insecure browsers during the U.S. crypto
1560 embargo. It requires special client support, and Exim is probably the
1561 only MTA that supported it -- and would never use it because real RSA is
1562 always available. This code has been removed, because it had the bad
1563 effect of slowing Exim down by computing (never used) parameters for the
1564 RSA_EXPORT functionality.
1566 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
1567 authenticator to fail if there's a tab character in the incoming data
1568 (there should never be unless someone is messing about, as it's supposed
1569 to be base64-encoded). Also added, on Timo's advice, the "secured" option
1570 if the connection is using TLS or if the remote IP is the same as the
1571 local IP, and the "valid-client-cert option" if a client certificate has
1574 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
1575 authenticators. This can be used for authorization after authentication
1576 succeeds. (In the case of plaintext, it servers for both authentication
1579 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
1580 if any retry times were supplied.
1582 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
1583 connection, where there is no client IP address to check. In this
1584 situation, the verify now always succeeds.
1586 PH/25 Applied John Jetmore's -Mset patch.
1588 PH/26 Added -bem to be like -Mset, but loading a message from a file.
1590 PH/27 In a string expansion for a processed (not raw) header when multiple
1591 headers of the same name were present, leading whitespace was being
1592 removed from all of them, but trailing whitespace was being removed only
1593 from the last one. Now trailing whitespace is removed from each header
1594 before concatenation. Completely empty headers in a concatenation (as
1595 before) are ignored.
1597 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
1598 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
1600 PH/29 [Removed. This was a change that I later backed out, and forgot to
1601 correct the ChangeLog entry (that I had efficiently created) before
1602 committing the later change.]
1604 PH/30 Exim was sometimes attempting to deliver messages that had suffered
1605 address errors (4xx response to RCPT) over the same connection as other
1606 messages routed to the same hosts. Such deliveries are always "forced",
1607 so retry times are not inspected. This resulted in far too many retries
1608 for the affected addresses. The effect occurred only when there were more
1609 hosts than the hosts_max_try setting in the smtp transport when it had
1610 the 4xx errors. Those hosts that it had tried were not added to the list
1611 of hosts for which the message was waiting, so if all were tried, there
1612 was no problem. Two fixes have been applied:
1614 (i) If there are any address or message errors in an SMTP delivery, none
1615 of the hosts (tried or untried) are now added to the list of hosts
1616 for which the message is waiting, so the message should not be a
1617 candidate for sending over the same connection that was used for a
1618 successful delivery of some other message. This seems entirely
1619 reasonable: after all the message is NOT "waiting for some host".
1620 This is so "obvious" that I'm not sure why it wasn't done
1621 previously. Hope I haven't missed anything, but it can't do any
1622 harm, as the worst effect is to miss an optimization.
1624 (ii) If, despite (i), such a delivery is accidentally attempted, the
1625 routing retry time is respected, so at least it doesn't keep
1626 hammering the server.
1628 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
1629 in ${readsocket because some servers need this prod.
1631 PH/32 Added some extra debug output when updating a wait-xxx database.
1633 PH/33 The hint "could be header name not terminated by colon", which has been
1634 given for certain expansion errors for a long time, was not being given
1635 for the ${if def:h_colon_omitted{... case.
1637 PH/34 The spec says: "With one important exception, whenever a domain list is
1638 being scanned, $domain contains the subject domain." There was at least
1639 one case where this was not true.
1641 PH/35 The error "getsockname() failed: connection reset by peer" was being
1642 written to the panic log as well as the main log, but it isn't really
1643 panic-worthy as it just means the connection died rather early on. I have
1644 removed the panic log writing for the ECONNRESET error when getsockname()
1647 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
1648 runs only) independently of the message's sender address. This meant
1649 that, if the 4xx error was in fact related to the sender, a different
1650 message to the same recipient with a different sender could confuse
1651 things. In particualar, this can happen when sending to a greylisting
1652 server, but other circumstances could also provoke similar problems.
1653 I have changed the default so that the retry time for these errors is now
1654 based a combination of the sender and recipient addresses. This change
1655 can be overridden by setting address_retry_include_sender=false in the
1658 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
1659 remote server are returned as part of bounce messages. This was not
1660 happening for LMTP over a pipe (the lmtp transport), but now it is the
1661 same for both kinds of LMTP.
1663 PH/38 Despite being documented as not happening, Exim was rewriting addresses
1664 in header lines that were in fact CNAMEs. This is no longer the case.
1666 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
1667 and queue runs started by the daemon processed all messages. This has
1668 been fixed so that -R and -S can now usefully be given with -q<time>.
1670 PH/40 Import PCRE release 6.7 (fixes some bugs).
1672 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
1674 PH/42 Give an error if -q is specified more than once.
1676 PH/43 Renamed the variables $interface_address and $interface_port as
1677 $received_ip_address and $received_port, to make it clear that these
1678 values apply to message reception, and not to the outgoing interface when
1679 a message is delivered. (The old names remain recognized, of course.)
1681 PH/44 There was no timeout on the connect() call when using a Unix domain
1682 socket in the ${readsocket expansion. There now is.
1684 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
1685 be meaningful with "accept".
1687 SC/01 Eximstats V1.43
1688 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
1690 SC/02 Eximstats V1.44
1691 Use a glob alias rather than an array ref in the generated
1692 parser. This improves both readability and performance.
1694 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
1695 Collect SpamAssassin and rejection statistics.
1696 Don't display local sender or destination tables unless
1697 there is data to show.
1698 Added average volumes into the top table text output.
1700 SC/04 Eximstats V1.46
1701 Collect data on the number of addresses (recipients)
1702 as well as the number of messages.
1704 SC/05 Eximstats V1.47
1705 Added 'Message too big' to the list of mail rejection
1706 reasons (thanks to Marco Gaiarin).
1708 SC/06 Eximstats V1.48
1709 Mainlog lines which have GMT offsets and are too short to
1710 have a flag are now skipped.
1712 SC/07 Eximstats V1.49 (Alain Williams)
1713 Added the -emptyok flag.
1715 SC/08 Eximstats V1.50
1716 Fixes for obtaining the IP address from reject messages.
1718 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
1719 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
1720 whitesspace changes from 4.64-PH/27
1722 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
1725 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
1726 are found, allow negative numbers in numeric criteria)
1728 JJ/06 exipick.20061117.2, added new $message_body_missing variable
1730 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
1731 to match changes made in 4.64-PH/43
1733 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
1735 PH/47 Put in an explicit test for a DNS lookup of an address record where the
1736 "domain" is actually an IP address, and force a failure. This locks out
1737 those revolvers/nameservers that support "A-for-A" lookups, in
1738 contravention of the specifications.
1740 PH/48 When a host name was looked up from an IP address, and the subsequent
1741 forward lookup of the name timed out, the host name was left in
1742 $sender_host_name, contrary to the specification.
1744 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
1745 restricted to single-key lookups, Exim was not diagnosing an error if
1746 * or *@ was used with a query-style lookup.
1748 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
1750 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
1751 long ago, but noone interested enough thought of it. Let's just say that
1752 the "1.1" means that there are some new functions that weren't there at
1753 some point in the past.
1755 PH/51 Error processing for expansion failure of helo_data from an smtp
1756 transport during callout processing was broken.
1758 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
1759 tested/used via the -bh/-bhc/-bs options.
1761 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
1762 bug, fixed in subsequent PCRE releases).
1764 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
1765 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
1767 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
1773 SC/01 Use a glob alias rather than an array ref in eximstats generated
1774 parser. This improves both readability and performance.
1776 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
1777 Don't display local sender or destination tables in eximstats unless
1778 there is data to show.
1779 Added average volumes into the eximstats top table text output.
1781 SC/03 Collect data on the number of addresses (recipients) as well
1782 as the number of messages in eximstats.
1784 TF/01 Correct an error in the documentation for the redirect router. Exim
1785 does (usually) call initgroups() when daemonizing.
1787 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
1788 with consistent privilege compared to when running as a daemon.
1790 TF/03 Note in the spec that $authenticated_id is not set for local
1791 submissions from trusted users.
1793 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
1794 Thanks to Dean Brooks <dean@iglou.com> for the patch.
1796 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
1797 by adding some example configuration directives to the default
1798 configuration file. A little bit of work is required to uncomment the
1799 directives and define how usernames and passwords are checked, but
1800 there is now a framework to start from.
1802 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
1803 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
1804 without this. I don't know how relevant this is to other LDAP libraries.
1806 PH/02 Add the verb name to the "unknown ACL verb" error.
1808 PH/03 Magnus Holmgren's patch for filter_prepend_home.
1810 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
1812 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
1813 directory not expanded when it should be if an expanded home directory
1814 was set for the address (which is overridden by the transport).
1816 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
1819 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
1820 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
1821 because it is too late at that time, and has no effect.
1823 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
1824 security issue with \' (bugzilla #107). I could not use the
1825 PQescapeStringConn() function, because it needs a PGconn value as one of
1828 PH/08 When testing addresses using -bt, indicate those final addresses that
1829 are duplicates that would not cause an additional delivery. At least one
1830 person was confused, thinking that -bt output corresponded to deliveries.
1831 (Suppressing duplicates isn't a good idea as you lose the information
1832 about possibly different redirections that led to the duplicates.)
1834 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
1835 systems where poll() doesn't work, in particular OS X.
1837 PH/10 Added more information to debugging output for retry time not reached.
1839 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
1840 operations in malware.c.
1842 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
1845 PH/13 If write_rejectlog was set false when logging was sent to syslog with
1846 syslog_duplication set false, log lines that would normally be written
1847 both the the main log and to the reject log were not written to syslog at
1850 PH/14 In the default configuration, change the use of "message" in ACL warn
1851 statements to "add_header".
1853 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
1854 not followed by a command (e.g. "seen endif").
1856 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
1857 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
1860 PH/17 Added extra conditions to the default value of delay_warning_condition
1864 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
1865 { match{$h_precedence:}{(?i)bulk|list|junk} } \
1866 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
1869 The Auto-Submitted: and various List- headers are standardised, whereas I
1870 don't think Precedence: ever was.
1872 PH/18 Refactored debugging code in route_finduser() to show more information,
1873 in particular, the error code if getpwnam() issues one.
1875 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
1876 This is apparently needed in addition to the PH/07 change above to avoid
1877 any possible encoding problems.
1879 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
1880 but not after initializing Perl.
1882 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
1883 output them only if debugging. By default they are written stderr,
1884 apparently, which is not desirable.
1886 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
1889 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
1892 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1894 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1895 authenticated or an ident call has been made. Suppress the default
1896 values for $authenticated_id and $authenticated_sender (but permit -oMai
1897 and -oMas) when testing with -bh.
1899 PH/24 Re-jigged the order of the tests in the default configuration so that the
1900 tests for valid domains and recipients precede the DNS black list and CSA
1901 tests, on the grounds that those ones are more expensive.
1903 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1904 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1905 This bug exists in every version of Exim that I still have, right back to
1908 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1909 However, an attempt to turn on case-sensitivity in a regex key by
1910 including (?-i) didn't work because the subject string was already
1911 lowercased, and the effects were non-intuitive. It turns out that a
1912 one-line patch can be used to allow (?-i) to work as expected.
1918 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1919 other effects) broke the use of negated acl sub-conditions.
1921 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1924 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1925 "Deny" causes Exim to reject the incoming connection with a 554 error.
1926 Unfortunately, if there is a major crisis, such as a disk failure,
1927 tcp-wrappers gives "deny", whereas what one would like would be some
1928 kind of temporary error. A kludge has been added to help with this.
1929 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1930 554 error is used if errno is still zero or contains ENOENT (which occurs
1931 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1934 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1936 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1937 errors. Otherwise a message that provokes a temporary error (when other
1938 messages do not) can cause a whole host to time out.
1940 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1941 addresses were routed directly to files or pipes from a redirect router.
1942 File deliveries just didn't batch; pipe deliveries might have suffered
1945 PH/06 A failure to get a lock for a hints database would erroneously always say
1946 "Failed to get write lock", even when it was really a read lock.
1948 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1949 of 0600. This has been changed to use the value of the lockfile_mode
1950 option (which defaults to 0600).
1952 PH/08 Applied small patch from the Sieve maintainer.
1954 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1955 folder from quota calculations, a direct delivery into this folder messed
1956 up the contents of the maildirsize file. This was because the regex was
1957 used only to exclude .Trash (or whatever) when the size of the mailbox
1958 was calculated. There was no check that a delivery was happening into an
1959 excluded directory. This bug has been fixed by ignoring all quota
1960 processing for deliveries into excluded directories.
1962 PH/10 Added the maildirfolder_create_regex option to appendfile.
1968 PH/01 The code for finding all the local interface addresses on a FreeBSD
1969 system running IPv6 was broken. This may well have applied to all BSD
1970 systems, as well as to others that have similar system calls. The broken
1971 code found IPv4 interfaces correctly, but gave incorrect values for the
1972 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1973 that it would not match correctly against @[] and not recognize the IPv6
1976 PH/02 The ipliteral router was not recognizing addresses of the form user@
1977 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1979 PH/03 Added disable_ipv6.
1981 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1982 decoded form, because it is most often used to construct To: headers
1983 lines in autoreplies, and the decoded form may well be syntactically
1984 invalid. However, $reply_address has leading white space removed, and all
1985 newlines turned into spaces so that the autoreply transport does not
1988 PH/05 If group was specified without a user on a router, and no group or user
1989 was specified on a transport, the group from the router was ignored.
1991 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1992 for visible compile-time settings that can be used to change these
1993 numbers, for those that want even more. Backwards compatibility with old
1994 spool files has been maintained. However, going back to a previous Exim
1995 release will lost any variables that are in spool files.
1997 PH/07 Two small changes when running in the test harness: increase delay when
1998 passing a TCP/IP connection to a new process, in case the original
1999 process has to generate a bounce, and remove special handling of
2000 127.0.0.2 (sic), which is no longer necessary.
2002 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2003 be the same on different OS.
2005 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2008 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2009 whether --show-vars was specified or not
2011 JJ/02 exipick: Added support for new ACL variable spool format introduced
2014 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2015 syntactically invalid From: or Reply-to: line, and a filter used this to
2016 generate an autoreply, and therefore failed to obtain an address for the
2017 autoreply, Exim could try to deliver to a non-existent relative file
2018 name, causing unrelated and misleading errors. What now happens is that
2019 it logs this as a hard delivery error, but does not attempt to create a
2022 PH/11 The exinext utility has a -C option for testing purposes, but although
2023 the given file was scanned by exinext itself; it wasn't being passed on
2024 when Exim was called.
2026 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2027 an end-of-file indication when reading a command response.
2029 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2030 compiled. In many other places in Exim, IPv6 addresses are always
2031 recognized, so I have changed this. It also means that IPv4 domain
2032 literals of the form [IPV4:n.n.n.n] are now always recognized.
2034 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2035 used if the router is not running as root, for example, when verifying at
2036 ACL time, or when using -bh. The debugging output from this situation was
2037 non-existent - all you got was a failure to exec. I have made two
2040 (a) Failures to set uid/gid, the current directory, or a process leader
2041 in a subprocess such as that created by queryprogram now generate
2042 suitable debugging ouput when -d is set.
2044 (b) The queryprogram router detects when it is not running as root,
2045 outputs suitable debugging information if -d is set, and then runs
2046 the subprocess without attempting to change uid/gid.
2048 PH/15 Minor change to Makefile for building test_host (undocumented testing
2051 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2052 additional section of a DNS packet that returns MX or SRV records.
2053 Instead, it always explicitly searches for A/AAAA records. This avoids
2054 major problems that occur when a DNS server includes only records of one
2055 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2056 fixed another bug: if SRV records were looked up and the corresponding
2057 address records were *not* found in the additional section, the port
2058 values from the SRV records were lost.
2060 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2061 using the correct key (the original address) when searching the retry
2062 rules in order to find which one to use for generating the retry hint.
2064 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2065 adding the default one. Similarly, if it contains a Reply-To: header, the
2066 errors_reply_to option, if set, is not used.
2068 PH/19 When calculating a retry time, Exim used to measure the "time since
2069 failure" by looking at the "first failed" field in the retry record. Now
2070 it does not use this if it is later than than the arrival time of the
2071 message. Instead it uses the arrival time. This makes for better
2072 behaviour in cases where some deliveries succeed, thus re-setting the
2073 "first failed" field. An example is a quota failure for a huge message
2074 when small messages continue to be delivered. Without this change, the
2075 "time since failure" will always be short, possible causing more frequent
2076 delivery attempts for the huge message than are intended.
2077 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2079 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2080 $1, $2, $3) because the numerical variables can be reset during some
2081 expansion items (e.g. "match"), thereby losing the authentication data.
2083 PH/21 Make -bV show the size of off_t variables so that the test suite can
2084 decide whether to run tests for quotas > 2G.
2086 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2087 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2088 filecount value is greater than 2G or if a quota value is greater than 2G
2089 on a system where the size of off_t is not greater than 4, a panic error
2092 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2093 never match. The debug and -bh output now contains an explicit error
2094 message indicating a malformed IPv4 address or mask.
2096 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2097 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2098 PH/23 above applies.
2100 PH/25 Do not write to syslog when running in the test harness. The only
2101 occasion when this arises is a failure to open the main or panic logs
2102 (for which there is an explicit test).
2104 PH/26 Added the /no_tell option to "control=freeze".
2106 PH/27 If a host name lookup failed very early in a connection, for example, if
2107 the IP address matched host_lookup and the reverse lookup yielded a name
2108 that did not have a forward lookup, an error message of the form "no IP
2109 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2110 could be logged. Now it outputs the IP address instead of "NULL".
2112 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2113 allows the sender and the authenticated sender to be set when
2114 submitting a message from within Exim. Since child_open_exim() is
2115 documented for local_scan(), the new function should be too.
2117 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2118 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2119 results in an empty string is now treated as unset.
2121 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2123 PH/31 Added sender_verify_fail logging option.
2125 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2126 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2127 tidied the source and removed it altogether.
2129 PH/33 When a queue run was abandoned because the load average was too high, a
2130 log line was always written; now it is written only if the queue_run log
2131 selector is set. In addition, the log line for abandonment now contains
2132 information about the queue run such as the pid. This is always present
2133 in "start" and "stop" lines but was omitted from the "abandon" line.
2135 PH/34 Omit spaces between a header name and the colon in the error message that
2136 is given when verify = headers_syntax fails (if there are lots of them,
2137 the message gets confusing).
2139 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2140 names, as there are now some PTR records that contain slashes. This check
2141 is only to protect against broken name servers that fall over on strange
2142 characters, so the fact that it applies to all lookups doesn't matter.
2144 PH/36 Now that the new test suite is complete, we can remove some of the
2145 special code in Exim that was needed for the old test suite. For example,
2146 sorting DNS records because real resolvers return them in an arbitrary
2147 order. The new test suite's fake resolver always returns records in the
2150 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2151 bounces) except when queue_only is set, to avoid logging races between
2152 the different processes.
2154 PH/38 Panic-die if .include specifies a non-absolute path.
2156 PH/39 A tweak to the "H" retry rule from its user.
2158 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2159 a label. They prevented compilation on older perls.
2161 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2162 a warning to be raised on newish perls.
2164 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2165 on queue. Changes to match documented behaviour of showing count of
2166 messages matching specified criteria.
2168 PH/40 Changed the default ident timeout from 30s to 5s.
2170 PH/41 Added support for the use of login_cap features, on those BSD systems
2171 that have them, for controlling the resources used by pipe deliveries.
2173 PH/42 The content-scanning code uses fopen() to create files in which to put
2174 message data. Previously it was not paying any attention to the mode of
2175 the files. Exim runs with umask(0) because the rest of the code creates
2176 files with open(), and sets the required mode explicitly. Thus, these
2177 files were ending up world-writeable. This was not a big issue, because,
2178 being within the spool directory, they were not world-accessible. I have
2179 created a function called modefopen, which takes an additional mode
2180 argument. It sets umask(777), creates the file, chmods it to the required
2181 mode, then resets the umask. All the relevant calls to fopen() in the
2182 content scanning code have been changed to use this function.
2184 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2185 to 24 hours. This avoids potential overflow problems when processing G
2186 and H retry rules. I suspect nobody ever tinkers with this value.
2188 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2190 PH/45 When the plaintext authenticator is running as a client, the server's
2191 challenges are checked to ensure they are valid base64 strings. By
2192 default, the authentication attempt is cancelled if an invalid string is
2193 received. Setting client_ignore_invalid_base64 true ignores these errors.
2194 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2195 they are received. Thus, the responses can be made to depend on the
2196 challenges. If an invalid string is ignored, an empty string is placed in
2199 PH/46 Messages that are created by the autoreply transport now contains a
2200 References: header, in accordance with RFCs 2822 and 3834.
2202 PH/47 Added authenticated_sender_force to the smtp transport.
2204 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2206 PH/49 Installed latest patch from the Sieve maintainer.
2208 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2209 was also set, the appendfile transport was unnecessarily scanning a
2210 directory of message files (e.g. for maildir delivery) to find the count
2211 of files (along with the size), even though it did not need this
2212 information. It now does the scan only if it needs to find either the
2213 size of the count of files.
2215 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2217 PH/52 Two bugs concerned with error handling when the smtp transport is
2220 (i) Exim was not creating retry information for temporary errors given
2221 for individual recipients after the DATA command when the smtp transport
2222 was used in LMTP mode. This meant that they could be retried too
2223 frequently, and not timed out correctly.
2225 (ii) Exim was setting the flag that allows error details to be returned
2226 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2227 recipients that were returned after the DATA command.
2229 PH/53 This is related to PH/52, but is more general: for any failing address,
2230 when detailed error information was permitted to be returned to the
2231 sender, but the error was temporary, then after the final timeout, only
2232 "retry timeout exceeded" was returned. Now it returns the full error as
2233 well as "retry timeout exceeded".
2235 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2236 do this, and (what is worse) MTAs that accept it.
2238 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2239 will now be deprecated.
2241 PH/56 New os.c-cygwin from the Cygwin maintainer.
2243 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2244 formats (previously only available in exim formats via -bpr, -bpru,
2245 and -bpra. Now also available in native and exiqgrep formats)
2247 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2248 with very large, slow to parse queues
2250 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2252 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2254 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2255 responses to authentication challenges, though it was showing the
2256 challenges; (ii) I've removed the CR characters from the debug output for
2259 PH/58 Allow for the insertion of a newline as well as a space when a string
2260 is turned into more than one encoded-word during RFC 2047 encoding. The
2261 Sieve code now uses this.
2263 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2264 data_4xx, lost_connection, tls_required.
2266 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2267 message was being sent as an SMTP response.
2269 PH/61 Add -l and -k options to exicyclog.
2271 PH/62 When verifying, if an address was redirected to one new address, so that
2272 verification continued, and the new address failed or deferred after
2273 having set something in $address_data, the value of $address_data was not
2274 passed back to the ACL. This was different to the case when no
2275 redirection occurred. The value is now passed back in both cases.
2277 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2278 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2279 use login_cap.h, so on its own it isn't the distinguishing feature. The
2280 new name refers directly to the setclassresources() function.
2282 PH/65 Added configuration files for NetBSD3.
2284 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2286 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2287 is preferred over IPv4.
2289 PH/68 The bounce_return_message and bounce_return_body options were not being
2290 honoured for bounces generated during the reception of non-SMTP messages.
2291 In particular, this applied to messages rejected by the ACL. This bug has
2292 been fixed. However, if bounce_return_message is true and bounce_return_
2293 body is false, the headers that are returned for a non-SMTP message
2294 include only those that have been read before the error was detected.
2295 (In the case of an ACL rejection, they have all been read.)
2297 PH/69 The HTML version of the specification is now built in a directory called
2298 spec_html instead of spec.html, because the latter looks like a path with
2299 a MIME-type, and this confuses some software.
2301 PH/70 Catch two compiler warnings in sieve.c.
2303 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2304 function verify_get_ident() calls ip_connect() to connect a socket, but
2305 if the "connect()" function timed out, ip_connect() used to close the
2306 socket. However, verify_get_ident() also closes the socket later, and in
2307 between Exim writes to the log, which may get opened at this point. When
2308 the socket was closed in ip_connect(), the log could get the same file
2309 descriptor number as the socket. This naturally causes chaos. The fix is
2310 not to close the socket in ip_connect(); the socket should be closed by
2311 the function that creates it. There was only one place in the code where
2312 this was missing, in the iplookup router, which I don't think anybody now
2313 uses, but I've fixed it anyway.
2315 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2316 well as to direct DNS lookups. Otherwise the handling of names in host
2317 lists is inconsistent and therefore confusing.
2323 PH/01 Two changes to the default runtime configuration:
2325 (1) Move the checks for relay_from_hosts and authenticated clients from
2326 after to before the (commented out) DNS black list checks.
2328 (2) Add control=submission to the relay_from_hosts and authenticated
2329 clients checks, on the grounds that messages accepted by these
2330 statements are most likely to be submissions.
2332 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2334 (1) Generate an error if the third argument for the ${prvs expansion is
2337 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2340 (3) Reset the variables that are obtained from the first argument of
2341 ${prvscheck and used in the second argument before leaving the code,
2342 because their memory is reclaimed, so using them afterwards may do
2345 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2346 one (it's much easier than Tom thought :-).
2348 (5) Because of (4), we can now allow for the use of $prvscheck_result
2349 inside the third argument.
2351 PH/03 For some reason, the default setting of PATH when running a command from
2352 a pipe transport was just "/usr/bin". I have changed it to
2355 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2356 anything to be listed in the output from -bV.
2358 PH/05 When a filter generated an autoreply, the entire To: header line was
2359 quoted in the delivery log line, like this:
2361 => >A.N.Other <ano@some.domain> <original@ddress> ...
2363 This has been changed so that it extracts the operative address. There
2364 may be more than one such address. If so, they are comma-separated, like
2367 => >ano@some.domain,ona@other.domain <original@ddress> ...
2369 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2370 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2371 not being looked up in the rDNS to get a host name, Exim was showing the
2372 IP address twice in Received: lines, even though the IP addresses were
2373 identical. For example:
2375 Received: from [1.2.3.4] (helo=[1.2.3.4])
2377 However, if the real host name was known, it was omitting the HELO data
2378 if it matched the actual IP address. This has been tidied up so that it
2379 doesn't show the same IP address twice.
2381 PH/07 When both +timestamp and +memory debugging was on, the value given by
2382 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2383 called by the debug printing, thereby overwriting the timestamp buffer.
2384 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2386 PH/08 When the original message was included in an autoreply transport, it
2387 always said "this is a copy of the message, including all the headers",
2388 even if body_only or headers_only was set. It now gives an appropriate
2391 PH/09 Applied a patch from the Sieve maintainer which:
2393 o fixes some comments
2394 o adds the (disabled) notify extension core
2395 o adds some debug output for the result of if/elsif tests
2396 o points to the current vacation draft in the documentation
2397 and documents the missing references header update
2401 o fixes a bug in processing the envelope test (when testing
2402 multiple envelope elements, the last element determinted the
2405 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2406 Electronic Mail") by including:
2408 Auto-submitted: auto-generated
2410 in the messages that it generates (bounce messages and others, such as
2411 warnings). In the case of bounce messages for non-SMTP mesages, there was
2412 also a typo: it was using "Auto_submitted" (underscore instead of
2413 hyphen). Since every message generated by Exim is necessarily in response
2414 to another message, thes have all been changed to:
2416 Auto-Submitted: auto-replied
2418 in accordance with these statements in the RFC:
2420 The auto-replied keyword:
2422 - SHOULD be used on messages sent in direct response to another
2423 message by an automatic process,
2425 - MUST NOT be used on manually-generated messages,
2427 - MAY be used on Delivery Status Notifications (DSNs) and Message
2428 Disposition Notifications (MDNs),
2430 - MUST NOT be used on messages generated by automatic or periodic
2431 processes, except for messages which are automatic responses to
2434 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2435 to the default Received: header definition.
2437 PH/12 Added log selector acl_warn_skipped (default on).
2439 PH/13 After a successful wildlsearch lookup, discard the values of numeric
2440 variables because (a) they are in the wrong storage pool and (b) even if
2441 they were copied, it wouldn't work properly because of the caching.
2443 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2444 checking when decoding. Apparently there are clients that generate
2445 overlong encoded strings. Why am I not surprised?
2447 PH/15 If the first argument of "${if match_address" was not empty, but did not
2448 contain an "@" character, Exim crashed. Now it writes a panic log message
2449 and treats the condition as false.
2451 PH/16 In autoreply, treat an empty string for "once" the same as unset.
2453 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
2454 extension "envelope-auth". The code is finished and in agreement with
2455 other implementations, but there is no documentation so far and in fact,
2456 nobody wrote the draft yet. This extension is currently #undef'ed, thus
2457 not changing the active code.
2459 Print executed "if" and "elsif" statements when debugging is used. This
2460 helps a great deal to understand what a filter does.
2462 Document more things not specified clearly in RFC3028. I had all this
2463 sorted out, when out of a sudden new issues came to my mind. Oops."
2465 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
2468 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
2469 canonical form (as documented). However, after a host name lookup from
2470 the IP address, check_host() was doing a simple string comparison with
2471 addresses acquired from the DNS when checking that the found name did
2472 have the original IP as one of its addresses. Since any found IPv6
2473 addresses are likely to be in abbreviated form, the comparison could
2474 fail. Luckily, there already exists a function for doing the comparison
2475 by converting both addresses to binary, so now that is used instead of
2476 the text comparison.
2478 PH/20 There was another similar case to PH/19, when a complete host name was
2479 given in a host list; looking up its IP address could give an abbreviated
2480 form, whereas the current host's name might or might not be abbreviated.
2481 The same fix has been applied.
2487 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
2488 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
2491 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
2492 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
2494 PH/03 Typo: missing ".o" in src/pcre/Makefile.
2496 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
2497 header line, restrict the check to what is listed in RFCs 2369 and 2929.
2498 Also, for "Auto-Submitted", treat anything other than "no" as
2499 non-personal, in accordance with RFC 3834. (Previously it treated
2500 anything starting "auto-" as non-personal.)
2502 TF/01 The control=submission/name=... option had a problem with syntax
2503 errors if the name included a slash character. The /name= option
2504 now slurps the rest of the string, so it can include any characters
2505 but it must come last in the list of options (after /sender_retain
2508 PH/05 Some modifications to the interface to the fake nameserver for the new
2516 TK/01 Added the "success_on_redirect" address verification option. See
2517 NewStuff for rationale and an example.
2519 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
2521 PH/02 Patch to exigrep to allow it to work on syslog lines.
2523 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
2524 fread() to skip over the body file's header line, because in Cygwin the
2525 header line is locked and is inaccessible.
2527 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
2528 co-exist for some time) to make it clear that it is the Exim ID that is
2529 referenced, not the Message-ID: header line.
2531 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
2532 string_format() function, because snprintf() does not exist on all
2535 PH/06 The use of forbid_filter_existstest now also locks out the use of the
2536 ${stat: expansion item.
2538 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
2539 protocol synchronization error", to keep the pedants happy.
2541 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
2542 well as for IRIX systems, when gcc is being used. See the host.c source
2545 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
2547 PH/10 Named domain lists were not working if used in a queue_smtp_domains
2550 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
2551 transport and to the smtp transport in LMTP mode.
2553 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
2555 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
2556 run a filter in a subprocess. This could lead to confusion in subsequent
2557 lookups in the parent process. There should also be a search_tidyup() at
2558 the end of the subprocess.
2560 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
2561 only if the host matched helo_try_verify_hosts, which caused the
2562 verification to occur when the EHLO/HELO command was issued. The ACL just
2563 tested the remembered result. Now, if a previous verification attempt has
2564 not happened, "verify = helo" does it there and then.
2566 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
2568 TK/03 Fix log output including CR from clamd.
2570 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
2571 exist provoked a memory error which could cause a segfault.
2573 PH/15 Installed PCRE 6.2
2575 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
2577 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
2578 of the problem. Specifically, suggested +O2 rather than +O1 for the
2581 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
2583 PH/20 If a delivery was routed to a non-standard port by means of an SRV
2584 record, the port was not correctly logged when the outgoing_port log
2585 selector was set (it logged the transort's default port).
2587 PH/21 Added support for host-specific ports to manualroute, queryprogram,
2588 fallback_hosts, and "hosts" in the smtp transport.
2590 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
2591 host errors such as "Connection refused".
2593 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
2594 authentication with radiusclient 0.4.9:
2596 - Error returned from rc_read_config was caught wrongly
2597 - Username/password not passed on to radius server due to wrong length.
2599 The presumption is that some radiusclient API changes for 4.51/PH/17
2600 were not taken care of correctly. The code is still untested by me (my
2601 Linux distribution still has 0.3.2 of radiusclient), but it was
2602 contributed by a Radius user.
2604 PH/24 When doing a callout, the value of $domain wasn't set correctly when
2605 expanding the "port" option of the smtp transport.
2607 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
2608 while reading a MIME header. Thanks to Tom Hughes for a patch.
2610 PH/24 Include config.h inside local_scan.h so that configuration settings are
2613 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
2614 that in an ACL for RCPT (for example), you can examine exactly what was
2617 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
2618 commands, but it was not correctly comparing the address with the actual
2619 client host address. Thus, it would show the EHLO address in Received:
2620 header lines when this was not necessary.
2622 PH/27 Added the % operator to ${eval:}.
2624 PH/28 Exim tries to create and chdir to its spool directory when it starts;
2625 it should be ignoring failures (because with -C, for example, it has lost
2626 privilege). It wasn't ignoring creation failures other than "already
2629 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
2632 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
2633 because an input line was too long, either on its own, or by virtue of
2634 too many continuations, the temporary file was not being removed, and the
2635 return code was incorrect.
2637 PH/31 Missing "BOOL" in function definition in filtertest.c.
2639 PH/32 Applied Sieve patches from the maintainer.
2641 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
2643 PH/33 Added "verify = not_blind".
2645 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
2646 Local/Makefile (with some defaults set). These are used in built scripts
2647 such as exicyclog, but they have never been used in the exim_install
2648 script (though there are many overriding facilities there). I have
2649 arranged that the exim_install script now takes note of these two
2652 PH/35 Installed configuration files for Dragonfly.
2654 PH/36 When a locally submitted message by a trusted user did not contain a
2655 From: header, and the sender address was obtained from -f or from an SMTP
2656 MAIL command, and the trusted user did not use -F to supply a sender
2657 name, $originator_name was incorrectly used when constructing a From:
2658 header. Furthermore, $originator_name was used for submission mode
2659 messages from external hosts without From: headers in a similar way,
2660 which is clearly wrong.
2662 PH/37 Added control=suppress_local_fixups.
2664 PH/38 When log_selector = +received_sender was set, and the addition of the
2665 sender made the log line's construction buffer exactly full, or one byte
2666 less than full, an overflow happened when the terminating "\n" was
2669 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
2670 when the result of a list match is failure because a DNS lookup failed.
2672 PH/40 RM_COMMAND is now used in the building process.
2674 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
2675 the "build-* directories that it finds.
2677 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
2678 address in a domain literal was a prefix of an interface address.
2680 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
2681 when verifying a sender address, unless rewrite_headers is false.
2683 PH/44 Wrote a long comment about why errors_to addresses are verified as
2684 recipients, not senders.
2686 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
2687 the ratelimit ACL was added.
2689 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
2691 PH/46 Added extra information about PostgreSQL errors to the error string.
2693 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
2694 avoiding the need to install special zones in a real server. This is
2695 backwards compatible; if it can't find the fake resolver, it drops back.
2696 Thus, both old and new test suites can be run.
2698 TF/02 Added util/ratelimit.pl
2700 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
2701 clock is set back in time.
2703 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
2704 Candler <B.Candler@pobox.com>.
2706 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
2707 OK for addresses that are the result of redirections.
2709 PH/48 A number of further additions for the benefit of the new test suite,
2710 including a fake gethostbyname() that interfaces to the fake DNS resolver
2713 TF/06 The fix for widen_domains has also been applied to qualify_single and
2714 search_parents which are the other dnslookup options that can cause
2717 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
2720 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
2722 TF/07 Exim produced the error message "an SRV record indicated no SMTP
2723 service" if it encountered an MX record with an empty target hostname.
2724 The message is now "an MX or SRV record indicated no SMTP service".
2726 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
2727 if the DNS of the sending site is misconfigured. This is quite a
2728 common situation. This change restores the behaviour of treating a
2729 helo verification defer as a failure.
2731 PH/51 If self=fail was set on a router, the bounce message did not include the
2732 actual error message.
2738 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
2740 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
2741 command itself ended in error, the underlying message about the transport
2742 filter timeout was being overwritten with the pipe command error. Now the
2743 underlying error message should be appended to the second error message.
2745 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
2747 PH/02 Reduce the amount of output that "make" produces by default. Full output
2748 can still be requested.
2750 PH/03 The warning log line about a condition test deferring for a "warn" verb
2751 was being output only once per connection, rather than after each
2752 occurrence (because it was using the same function as for successful
2753 "warn" verbs). This seems wrong, so I have changed it.
2755 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
2756 it should not have, which might have caused a crash in the right
2757 circumstances, but probably never did.
2759 PH/04 Installed a modified version of Tony Finch's patch to make submission
2760 mode fix the return path as well as the Sender: header line, and to
2761 add a /name= option so that you can make the user's friendly name appear
2764 TF/03 Added the control = fakedefer ACL modifier.
2766 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
2767 Mark Lowes for thorough testing.
2769 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
2771 TK/03 Merged latest SRS patch from Miles Wilton.
2773 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
2774 with the definition in sysexits.h (which is #included earlier).
2775 Fortunately, Exim does not actually use EX_OK. The code used to try to
2776 preserve the sysexits.h value, by assumimg that macro definitions were
2777 scanned for macro replacements. I have been disabused of this notion,
2778 so now the code just undefines EX_OK before #including unistd.h.
2780 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
2781 in the smtp transport. When a block could not be written in a single
2782 write() function, the timeout was being re-applied to each part-write.
2783 This seems wrong - if the receiver was accepting one byte at a time it
2784 would take for ever. The timeout is now adjusted when this happens. It
2785 doesn't have to be particularly precise.
2787 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
2788 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
2790 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
2791 without a domain if the check to <postmaster@domain> fails.
2793 SC/01 Eximstats: added -xls and the ability to specify output files
2794 (patch written by Frank Heydlauf).
2796 SC/02 Eximstats: use FileHandles for outputing results.
2798 SC/03 Eximstats: allow any combination of xls, txt, and html output.
2800 SC/04 Eximstats: fixed display of large numbers with -nvr option
2802 SC/05 Eximstats: fixed merging of reports with empty tables.
2804 SC/06 Eximstats: added the -include_original_destination flag
2806 SC/07 Eximstats: removed tabs and trailing whitespace.
2808 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
2810 TK/06 MBOX spool code: Add real "From " MBOX separator line
2811 so the .eml file is really in mbox format (even though
2812 most programs do not really care). Patch from Alex Miller.
2814 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
2815 The latter is generated from $received_to and is only set if the
2816 message has one envelope recipient. SA can use these headers,
2817 obviously out-of-the-box. Patch from Alex Miller.
2819 PH/08 The ${def test on a variable was returning false if the variable's
2820 value was "0", contrary to what the specification has always said!
2821 The result should be true unless the variable is empty.
2823 PH/09 The syntax error of a character other than { following "${if
2824 def:variable_name" (after optional whitespace) was not being diagnosed.
2825 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
2826 accidental colon was present, for example, could give incorrect results.
2828 PH/10 Tidied the code in a number of places where the st_size field of a stat()
2829 result is used (not including appendfile, where other changes are about
2832 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
2833 This involved changing a lot of size variables from int to off_t. It
2834 should work with maildirs and everything.
2836 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
2837 spamd dying while we are connected to it.
2839 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
2842 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
2843 function for BDB 4.3.
2845 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
2847 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
2848 the sources for PCRE have been re-arranged and more files are now
2851 PH/15 The code I had for printing potentially long long variables in PH/11
2852 above was not the best (it lost precision). The length of off_t variables
2853 is now inspected at build time, and an appropriate printing format (%ld
2854 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
2855 to be "long long int" or "long int". This is needed for the internal
2856 formatting function string_vformat().
2858 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
2859 the configuration file to be ":syslog", then the script "guesses" where
2860 the logs files are, rather than using the compiled in default. In our
2861 case the guess is not the same as the compiled default, so the script
2862 suddenly stopped working when I started to use syslog. The patch checks
2863 to see if log_file_path is "". If so, it attempts to read it from exim
2864 with no configuration file to get the compiled in version, before it
2865 falls back to the previous guessing code."
2867 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
2868 implementing BATV in an Exim configuration. See NewStuff for the gory
2871 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
2872 Makefile that are specific to HP-UX.
2874 PH/18 If the "use_postmaster" option was set for a recipient callout together
2875 with the "random" option, the postmaster address was used as the MAIL
2876 FROM address for the random test, but not for the subsequent recipient
2877 test. It is now used for both.
2879 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
2880 patch removes a few documentation additions to RFC 3028, because the
2881 latest draft now contains them. It adds the new en;ascii-case comparator
2882 and a new error check for 8bit text in MIME parts. Comparator and
2883 require names are now matched exactly. I enabled the subaddress
2884 extension, but it is not well tested yet (read: it works for me)."
2886 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
2887 rework some of the code of TK/09 above to avoid the hardwired use of
2888 "%lld" and "long long". Replaced the call to snprintf() with a call to
2891 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
2892 records point to non-existent hosts", "retry timeout exceeded", and
2893 "retry time not reached for any host after a long failure period".
2895 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2896 experimental DomainKeys support:
2898 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2899 (2) On an error such as an illegally used "control", the wrong name for
2900 the control was given.
2902 These problems did NOT occur unless DomainKeys support was compiled.
2904 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2906 PH/24 Added ${if match_ip condition.
2908 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2909 databases so that it will be absolutely obvious if a crash occurs in the
2910 DB library. This is a regular occurrence (often caused by mis-matched
2913 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2914 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2915 picked up on a user's system that detects such things. There doesn't seem
2916 to be a gcc warning option for this - only an attribute that has to be
2917 put on the function's prototype. It seems that in Fedora Core 4 they have
2918 set this on a number of new functions. No doubt there will be more in due
2921 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2922 specify a transport. However, if an address that was verified by such a
2923 router was the subject of a callout, Exim crashed because it tried to
2924 read the rcpt_include_affixes from the non-existent transport. Now it
2925 just assumes that the setting of that option is false. This bug was
2926 introduced by 4.51/PH/31.
2928 PH/28 Changed -d+all to exclude +memory, because that information is very
2929 rarely of interest, but it makes the output a lot bigger. People tend to
2930 do -d+all out of habit.
2932 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2933 code in os-type was giving problems when libc.so lives in lib64, like on
2936 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2937 aren't the modern standard, and it seems that some systems' include files
2938 don't always have them. Exim was already checking for some of the newer
2939 ones like T_AAAA, and defining it itself. I've added checks for all the
2940 record types that Exim uses.
2942 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2943 not automatically generating a new one, as it is supposed to. This
2944 prevented TLS from working. If the file did exist, but contained invalid
2945 data, a new version was generated, as expected. It was only the case of a
2946 non-existent file that was broken.
2948 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2949 with a change in libdomainkeys > 0.64.
2951 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2952 from DNS. If the selector record carries the flag, it now has
2953 precedence over the domain-wide flag.
2955 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2957 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2958 the use of a port name that isn't defined in /etc/services) occurred, the
2959 message was deferred as in a normal delivery, and thus remained on the
2960 spool, instead of being failed because of the mua_wrapper setting. This
2961 is now fixed, and I tidied up some of the mua_wrapper messages at the
2964 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2965 the messages in a hash of arrays rather than using individual hashes.
2966 This is a bit cleaner and results in dramatic memory savings, albeit
2967 at a slight CPU cost.
2969 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2970 as requested by Marc Sherman.
2972 SC/10 Eximstats: added histograms for user specified patterns as requested
2975 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2977 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2978 fopen() in the content-scanning modules that did not already have it.
2984 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2985 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2987 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2989 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2991 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2992 to test Sieve filters that use "vacation".
2994 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2995 that changes the way the GnuTLS parameters are stored in the cache file.
2996 The new format can be generated externally. For backward compatibility,
2997 if the data in the cache doesn't make sense, Exim assumes it has read an
2998 old-format file, and it generates new data and writes a new file. This
2999 means that you can't go back to an older release without removing the
3002 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3003 work if there are any delivery delays because "one_time" forces the
3004 parent to be marked "delivered", so its unseen clone is never tried
3005 again. For this reason, Exim now forbids the simultaneous setting of
3008 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3009 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3010 entry if you want to know the details. The fix, however, neglected to
3011 consider the case when local delivery batching is involved. The test for
3012 "previously delivered" was not happening when checking to see if an
3013 address could be batched with a previous (undelivered) one; under
3014 certain circumstances this could lead to multiple deliveries to the same
3017 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3018 in its include files, and this causes problems building Exim.
3020 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3021 header_syntax) but Exim was just ignoring anything given after a slash.
3022 In particular, this caused confusion with an attempt to use "verify =
3023 reverse_host_lookup/defer_ok". An error is now given when options are
3024 supplied for verify items that do not have them. (Maybe reverse_host_
3025 lookup should have a defer_ok option, but that's a different point.)
3027 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3028 defined by RFC 821) to 2048, because there were problems with some AUTH
3029 commands, and RFC 1869 says the size should be increased for extended
3030 SMTP commands that take arguments.
3032 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3035 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3036 "unknown" error; now it says that the functionality isn't in the binary.
3038 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3039 an address' error message when a string expansion fails (syntax or
3040 whatever). Otherwise the password may appear in the log. Following change
3041 PH/42 below, there is no longer a chance of it appearing in a bounce
3044 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3046 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3047 of its addresses was ever tried. (Bugzilla bug #2.)
3049 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3050 the result incorrectly in the debug output. (It correctly added a newline
3051 to what was transported.)
3053 TF/01 Added $received_time.
3055 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3056 commented out examples of how to interface to a virus scanner and to
3057 SpamAssassin. Also added commented examples of av_scanner and
3058 spamd_address settings.
3060 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3061 and controls are allowed in which ACLs. There were a couple of minor
3062 errors. Some of the entries in the conditions table (which is a table of
3063 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3064 negation of where the condition IS allowed.
3066 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3068 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3069 header file does not have a version number, so I've had to invent a new
3070 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3071 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3072 radiusclient), but it was contributed by a Radius user.
3074 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3075 files or directories, for OpenSSL.
3077 PH/19 When an Exim process that is running as root has to create an Exim log
3078 file, it does so in a subprocess that runs as exim:exim so as to get the
3079 ownership right at creation (otherwise, other Exim processes might see
3080 the file with the wrong ownership). There was no test for failure of this
3081 fork() call, which would lead to the process getting stuck as it waited
3082 for a non-existent subprocess. Forks do occasionally fail when resources
3083 run out. I reviewed all the other calls to fork(); they all seem to check
3086 PH/20 When checking for unexpected SMTP input at connect time (before writing
3087 the banner), Exim was not dealing correctly with a non-positive return
3088 from the read() function. If the client had disconnected by this time,
3089 the result was a log entry for a synchronization error with an empty
3090 string after "input=" when read() returned zero. If read() returned -1
3091 (an event I could not check), uninitialized data bytes were printed.
3092 There were reports of junk text (parts of files, etc) appearing after
3095 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3097 PH/22 Added support for macro redefinition, and (re)definition in between
3098 driver and ACL definitions.
3100 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3101 forgetting to use the resulting value; it was using the unexpanded value.
3103 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3104 hadn't been configured. The fix is from Juergen Kreileder, who
3105 understands it better than I do:
3107 "Here's what I see happening with three configured cyrus_sasl
3108 authenticators configured (plain, login, cram-md5):
3110 On startup auth_cyrus_sasl_init() gets called for each of these.
3111 This means three calls to sasl_listmech() without a specified mech_list.
3112 => SASL tests which mechs of all available mechs actually work
3113 => three warnings about OTP not working
3114 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3116 With the patch, sasl_listmech() also gets called three times. But now
3117 SASL's mech_list option is set to the server_mech specified in the the
3118 authenticator. Or in other words, the answer from sasl_listmech()
3119 gets limited to just the mech you're testing for (which is different
3121 => the return list contains just 'plain' or 'login', 'cram-md5' or
3122 nothing depending on the value of ob->server_mech.
3124 I've just tested the patch: Authentication still works fine,
3125 unavailable mechs specified in the exim configuration are still
3126 caught, and the auth.log warnings about OTP are gone."
3128 PH/25 When debugging is enabled, the contents of the command line are added
3129 to the debugging output, even when log_selector=+arguments is not
3132 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3133 answer is "GNU", and only if the return is "GNU/something" is the answer
3136 PH/27 $acl_verify_message is now set immediately after the failure of a
3137 verification in an ACL, and so is available in subsequent modifiers. In
3138 particular, the message can be preserved by coding like this:
3140 warn !verify = sender
3141 set acl_m0 = $acl_verify_message
3143 Previously, $acl_verify_message was set only while expanding "message"
3144 and "log_message" when a very denied access.
3146 PH/28 Modified OS/os.c-Linux with
3148 -#ifndef OS_LOAD_AVERAGE
3149 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3151 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3152 nomenclature these days.)
3154 PH/29 Installed patch from the Sieve maintainer that adds the options
3155 sieve_useraddress and sieve_subaddress to the redirect router.
3157 PH/30 In these circumstances:
3158 . Two addresses routed to the same list of hosts;
3159 . First host does not offer TLS;
3160 . First host accepts first address;
3161 . First host gives temporary error to second address;
3162 . Second host offers TLS and a TLS session is established;
3163 . Second host accepts second address.
3164 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3165 and peerdn, if requested) that were in fact used only for the second
3168 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3169 attention to any local part prefix or suffix that was matched by the
3170 router that accepted the address. It now behaves in the same way as it
3171 does for delivery: the affixes are removed from the local part unless
3172 rcpt_include_affixes is set on the transport.
3174 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3175 timeout during the DATA phase of an incoming message.
3177 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3178 applied a patch sanctioned by the Sieve maintainer.
3180 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3181 the uid or gid is negative. A case of a negative gid caused this to be
3182 noticed. The fix allows for either to be negative.
3184 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3185 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3188 PH/36 The change PH/12 above was broken. Fixed it.
3190 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3191 the grounds that routing the same address twice would always produce the
3192 same answer. This might have been true once, but it is certainly no
3193 longer true now. Routing a child address may depend on the previous
3194 routing that produced that child. Some complicated redirection strategies
3195 went wrong when messages had multiple recipients, and made Exim's
3196 behaviour dependent on the order in which the addresses were given.
3198 I have moved the duplicate checking until after the routing is complete.
3199 Exim scans the addresses that are assigned to local and remote
3200 transports, and removes any duplicates. This means that more work will be
3201 done, as duplicates will always all be routed, but duplicates are
3202 presumably rare, so I don't expect this is of any significance.
3204 For deliveries to pipes, files, and autoreplies, the duplicate checking
3205 still happens during the routing process, since they are not going to be
3208 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3209 It corrects a timeout issue with spamd. This is Ian's comment: "The
3210 background is that sometimes spamd either never reads data from a
3211 connection it has accepted, or it never writes response data. The exiscan
3212 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3213 blindly assumes that writes won't block so it may never time out."
3215 PH/39 Allow G after quota size as well as K and M.
3217 PH/40 The value set for $authenticated_id in an authenticator may not contain
3218 binary zeroes or newlines because the value is written to log lines and
3219 to spool files. There was no check on this. Now the value is run through
3220 the string_printing() function so that such characters are converted to
3221 printable escape sequences.
3223 PH/41 $message_linecount is a new variable that contains the total number of
3224 lines in the message. Compare $body_linecount, which is the count for the
3227 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3228 bounce and delay warning messages, except in certain special cases, which
3231 (a) An SMTP error message from a remote host;
3232 (b) A message specified in a :fail: redirection;
3233 (c) A message specified in a "fail" command in a system filter;
3234 (d) A message specified in a FAIL return from the queryprogram router;
3235 (e) A message specified by the cannot_route_message router option.
3237 In these cases only, Exim does include the error details in bounce and
3238 warning messages. There are also a few cases where bland messages such
3239 as "unrouteable address" or "local delivery error" are given.
3241 PH/43 $value is now also set for the "else" part of a ${run expansion.
3243 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3244 being worked on, but at least Exim now implements the latest version to
3247 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3248 process to complete was treated as a delivery failure, a timeout while
3249 writing the message to the pipe was logged, but erroneously treated as a
3250 successful delivery. Such timeouts include transport filter timeouts. For
3251 consistency with the overall process timeout, these timeouts are now
3252 treated as errors, giving rise to delivery failures by default. However,
3253 there is now a new Boolean option for the pipe transport called
3254 timeout_defer, which, if set TRUE, converts the failures into defers for
3255 both kinds of timeout. A transport filter timeout is now identified in
3258 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3259 systems where "make" and "gmake" are different, calling "gmake" at top
3260 level broke things. I've arranged for the value of $(MAKE) to be passed
3261 from the Makefile to this script so that it can call the same version of
3265 A note about Exim versions 4.44 and 4.50
3266 ----------------------------------------
3268 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3269 changes of various kinds. As a consequence, a big documentation update was
3270 needed. This delayed the release for rather longer than seemed good, especially
3271 in the light of a couple of (minor) security issues. Therefore, the changes
3272 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3273 release. So 4.44 and 4.50 are in effect two different branches that both start
3276 I have left the 4.50 change log unchanged; it contains all the changes since
3277 4.43. The change log for 4.44 is below; many of its items are identical to
3278 those for 4.50. This seems to be the most sensible way to preserve the
3279 historical information.
3285 1. Minor wording change to the doc/README.SIEVE file.
3287 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3288 computation of the current number of files was incorrect.
3290 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3291 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3294 4. Give more explanation in the error message when the command for a transport
3295 filter fails to execute.
3297 5. There are several places where Exim runs a non-Exim command in a
3298 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3299 was being done only for the command run by the queryprogram router. It is
3300 now done for all such subprocesses. The other cases are: ${run, transport
3301 filters, and the commands run by the lmtp and pipe transports.
3303 6. Added CONFIGURE_GROUP build-time option.
3305 7. Some older OS have a limit of 256 on the maximum number of file
3306 descriptors. Exim was using setrlimit() to set 1000 as a large value
3307 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3308 systems. I've change it so that if it can't get 1000, it tries for 256.
3310 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3311 was an oversight, and furthermore, ever since the addition of extra
3312 controls (e.g. 4.43/32), the checks on when to allow different forms of
3313 "control" were broken. There should now be diagnostics for all cases when a
3314 control that does not make sense is encountered.
3316 9. Added the /retain_sender option to "control=submission".
3318 10. $recipients is now available in the predata ACL (oversight).
3320 11. Tidy the search cache before the fork to do a delivery from a message
3321 received from the command line. Otherwise the child will trigger a lookup
3322 failure and thereby defer the delivery if it tries to use (for example) a
3323 cached ldap connection that the parent has called unbind on.
3325 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3326 of $address_data from the recipient verification was clobbered by the
3327 sender verification.
3329 13. The value of address_data from a sender verification is now available in
3330 $sender_address_data in subsequent conditions in the ACL statement.
3332 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3334 15. Added a new option "connect=<time>" to callout options, to set a different
3337 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3338 was its contents. (It was OK if the option was not defined at all.)
3340 17. A "Completed" log line is now written for messages that are removed from
3341 the spool by the -Mrm option.
3343 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3344 information about exactly what failed.
3346 19. Added -dd to debug only the daemon process.
3348 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3349 handles timeouts, both on the server side and network timeouts. Renamed the
3350 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3352 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3353 It is now set to "smtps".
3355 22. $host_address is now set to the target address during the checking of
3356 ignore_target_hosts.
3358 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3359 being passed; this would have caused $sender_host_name to have been used if
3360 matching the list had actually called for a host name (not very likely,
3361 since this list is usually IP addresses). A host name is now passed as
3364 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3365 code that specifies a non-restarting handler (typically sigaction() in
3366 modern systems) in an attempt to fix a rare and obscure crash bug.
3368 25. Narrowed the window for a race in the daemon that could cause it to ignore
3369 SIGCHLD signals. This is not a major problem, because they are used only to
3370 wake it up if nothing else does.
3372 26. A malformed maildirsize file could cause Exim to calculate negative values
3373 for the mailbox size or file count. Odd effects could occur as a result.
3374 The maildirsize information is now recalculated if the size or filecount
3377 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3378 support for a long time. Removed HAVE_SYS_VFS_H.
3380 28. Installed the latest version of exipick from John Jetmore.
3382 29. In an address list, if the pattern was not a regular expression, an empty
3383 subject address (from a bounce message) matched only if the pattern was an
3384 empty string. Non-empty patterns were not even tested. This was the wrong
3385 because it is perfectly reasonable to use an empty address as part of a
3386 database query. An empty address is now tested by patterns that are
3387 lookups. However, all the other forms of pattern expect the subject to
3388 contain a local part and a domain, and therefore, for them, an empty
3389 address still always fails if the pattern is not itself empty.
3391 30. Exim went into a mad DNS loop when attempting to do a callout where the
3392 host was specified on an smtp transport, and looking it up yielded more
3393 than one IP address.
3395 31. Re-factored the code for checking spool and log partition space into a
3396 function that finds that data and another that does the check. The former
3397 is then used to implement four new variables: $spool_space, $log_space,
3398 $spool_inodes, and $log_inodes.
3400 32. The RFC2047 encoding function was originally intended for short strings
3401 such as real names; it was not keeping to the 75-character limit for
3402 encoded words that the RFC imposes. It now respects the limit, and
3403 generates multiple encoded words if necessary. To be on the safe side, I
3404 have increased the buffer size for the ${rfc2047: expansion operator from
3407 33. It is now permitted to omit both strings after an "if" condition; if the
3408 condition is true, the result is "true". As before, when the second string
3409 is omitted, a false condition yields an empty string. This makes it less
3410 cumbersome to write custom ACL and router conditions.
3412 34. Failure to deliver a bounce message always caused it to be frozen, even if
3413 there was an errors_to setting on the router. The errors_to setting is now
3416 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3417 canonical form (fully expanded) before being placed in
3418 $sender_host_address.
3420 36. The table in the code that translates DNS record types into text (T_A to
3421 "A" for instance) was missing entries for NS and CNAME. It is just possible
3422 that this could have caused confusion if both these types were looked up
3423 for the same domain, because the text type is used as part of Exim's
3424 per-process caching. But the chance of anyone hitting this buglet seems
3427 37. The dnsdb lookup has been extended in a number of ways.
3429 (1) There is a new type, "zns", which walks up the domain tree until it
3430 finds some nameserver records. It should be used with care.
3432 (2) There is a new type, "mxh", which is like "mx" except that it returns
3433 just the host names, not the priorities.
3435 (3) It is now possible to give a list of domains (or IP addresses) to be
3436 looked up. The behaviour when one of the lookups defers can be
3437 controlled by a keyword.
3439 (4) It is now possible to specify the separator character for use when
3440 multiple records are returned.
3442 38. The dnslists ACL condition has been extended: it is now possible to supply
3443 a list of IP addresses and/or domains to be looked up in a particular DNS
3446 39. Added log_selector=+queue_time_overall.
3448 40. When running the queue in the test harness, wait just a tad after forking a
3449 delivery process, to get repeatability of debugging output.
3451 41. Include certificate and key file names in error message when GnuTLS fails
3452 to set them up, because the GnuTLS error message doesn't include the name
3453 of the failing file when there is a problem reading it.
3455 42. Allow both -bf and -bF in the same test run.
3457 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
3459 44. The "Exiscan patch" is now merged into the mainline Exim source.
3461 45. Sometimes the final signoff response after QUIT could fail to get
3462 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3463 < 0 before doing a fflush(). This bug looks as though it goes back to the
3464 introduction of TLS in release 3.20, but "sometimes" must have been rare
3465 because the tests only now provoked it.
3467 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
3468 (this can affect the format of dates).
3470 47. exim_tidydb, when checking for the continued existence of a message for
3471 which it has found a message-specific retry record, was not finding
3472 messages that were in split spool directories. Consequently, it was
3473 deleting retry records that should have stayed in existence.
3475 48. Steve fixed some bugs in eximstats.
3477 49. The SPA authentication driver was not abandoning authentication and moving
3478 on to the next authenticator when an expansion was forced to fail,
3479 contradicting the general specification for all authenticators. Instead it
3480 was generating a temporary error. It now behaves as specified.
3482 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
3483 (the order specifies the preference for clients). The order is now AES256,
3484 AES128, 3DES, ARCFOUR128.
3486 51. Small patch to Sieve code - explicitly set From: when generating an
3489 52. Exim crashed if a remote delivery caused a very long error message to be
3490 recorded - for instance if somebody sent an entire SpamAssassin report back
3491 as a large number of 550 error lines. This bug was coincidentally fixed by
3492 increasing the size of one of Exim's internal buffers (big_buffer) that
3493 happened as part of the Exiscan merge. However, to be on the safe side, I
3494 have made the code more robust (and fixed the comments that describe what
3497 53. Now that there can be additional text after "Completed" in log lines (if
3498 the queue_time_overall log selector is set), a one-byte patch to exigrep
3499 was needed to allow it to recognize "Completed" as not the last thing in
3502 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
3503 patch that reportedly fixes this has been added. I am not expert enough to
3504 create a test for it. This is what the patch creator wrote:
3506 "I found a little strange behaviour of ldap code when working with
3507 Windows 2003 AD Domain, where users was placed in more than one
3508 Organization Units. When I tried to give exim partial DN, the exit code
3509 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
3510 But simultaneously result of request was absolutely normal ldap result,
3511 so I produce this patch..."
3513 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
3514 REFERENCE, so I have modified the code to exclude the patch when that macro
3517 55. Some experimental protocols are using DNS PTR records for new purposes. The
3518 keys for these records are domain names, not reversed IP addresses. The
3519 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3520 leaves it alone. Component reversal etc. now happens only for IP addresses.
3523 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
3525 57. Double the size of the debug message buffer (to 2048) so that more of very
3526 long debug lines gets shown.
3528 58. The exicyclog utility now does better if the number of log files to keep
3529 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
3531 59. Two changes related to the smtp_active_hostname option:
3533 (1) $smtp_active_hostname is now available as a variable.
3534 (2) The default for smtp_banner uses $smtp_active_hostname instead
3535 of $primary_hostname.
3537 60. The host_aton() function is supposed to be passed a string that is known
3538 to be a valid IP address. However, in the case of IPv6 addresses, it was
3539 not checking this. This is a hostage to fortune. Exim now panics and dies
3540 if the condition is not met. A case was found where this could be provoked
3541 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3542 components; fortuitously, this particular loophole had already been fixed
3543 by change 4.50/55 above.
3545 If there are any other similar loopholes, the new check in host_aton()
3546 itself should stop them being exploited. The report I received stated that
3547 data on the command line could provoke the exploit when Exim was running as
3548 exim, but did not say which command line option was involved. All I could
3549 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3550 running as the user.
3553 61. There was a buffer overflow vulnerability in the SPA authentication code
3554 (which came originally from the Samba project). I have added a test to the
3555 spa_base64_to_bits() function which I hope fixes it.
3558 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
3559 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
3561 63. The daemon start-up calls getloadavg() while still root for those OS that
3562 need the first call to be done as root, but it missed one case: when
3563 deliver_queue_load_max is set with deliver_drop_privilege. This is
3564 necessary for the benefit of the queue runner, because there is no re-exec
3565 when deliver_drop_privilege is set.
3567 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
3568 This has been fixed.
3570 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3571 was in use, was not putting the data itself into the right store pool;
3572 consequently, it could be overwritten for a subsequent message in the same
3573 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3576 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
3578 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
3579 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
3580 the function were treating the return as a boolean value, which happened to
3581 work because 0=false and not-0=true, but is not correct code.
3583 68. The host_aton() function was not handling scoped IPv6 addresses (those
3584 with, for example, "%eth0" on the end) correctly.
3586 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
3587 negated items (that is, ~something) in unsigned ints. Some compilers
3588 apparently mutter when there is no cast.
3590 70. If an address verification called from an ACL failed, and did not produce a
3591 user-specific message (i.e. there was only a "system" message), nothing was
3592 put in $acl_verify_message. In this situation, it now puts the system
3595 71. Change 4.23/11 added synchronization checking at the start of an SMTP
3596 session; change 4.31/43 added the unwanted input to the log line - except
3597 that it did not do this in the start of session case. It now does.
3599 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
3600 This is wrong and can cause the other end to generate a synchronization
3601 error if it is another Exim or anything else that does the synchronization
3602 check. A QUIT command is no longer sent after a timeout.
3604 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
3605 during host lookups.
3607 74. The defer_ok option of callout verification was not working if it was used
3608 when verifying addresses in header lines, that is, for this case:
3610 verify = header_sender/callout=defer_ok
3612 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
3613 those file descriptors could be used for SMTP connections. If anything
3614 wrote to stderr (the example that came up was "warn" in embedded Perl), it
3615 could be sent to the SMTP client, causing chaos. The daemon now opens
3616 stdin, stdout, and stderr to /dev/null when it puts itself into the
3619 76. Arrange for output from Perl's "warn" command to be written to Exim's main
3620 log by default. The user can override this with suitable Perl magic.
3622 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
3623 the log message when discard triggers, was not working for the DATA ACL or
3624 for the non-SMTP ACL.
3626 78. Error message wording change in sieve.c.
3628 79. If smtp_accept_max_per_host was set, the number of connections could be
3629 restricted to fewer than expected, because the daemon was trying to set up
3630 a new connection before checking whether the processes handling previous
3631 connections had finished. The check for completed processes is now done
3632 earlier. On busy systems, this bug wouldn't be noticed because something
3633 else would have woken the daemon, and it would have reaped the completed
3636 80. If a message was submitted locally by a user whose login name contained one
3637 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
3638 It caused a spool format error. I have fixed the spool reading code. A
3639 related problem was that the "from" clause in the Received: line became
3640 illegal because of the space(s). It is now covered by ${quote_local_part.
3642 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
3645 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
3647 83. Patch from Sieve maintainer to fix unterminated string problem in
3648 "vacation" handling.
3650 84. Some minor changes to the Linux configuration files to help with other
3651 OS variants using glibc.
3653 85. One more patch for Sieve to update vacation handling to latest spec.
3656 ----------------------------------------------------
3657 See the note above about the 4.44 and 4.50 releases.
3658 ----------------------------------------------------
3664 1. Change 4.43/35 introduced a bug that caused file counts to be
3665 incorrectly computed when quota_filecount was set in an appendfile
3668 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3669 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3672 3. Give more explanation in the error message when the command for a transport
3673 filter fails to execute.
3675 4. There are several places where Exim runs a non-Exim command in a
3676 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3677 was being done only for the command run by the queryprogram router. It is
3678 now done for all such subprocesses. The other cases are: ${run, transport
3679 filters, and the commands run by the lmtp and pipe transports.
3681 5. Some older OS have a limit of 256 on the maximum number of file
3682 descriptors. Exim was using setrlimit() to set 1000 as a large value
3683 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3684 systems. I've change it so that if it can't get 1000, it tries for 256.
3686 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3687 was an oversight, and furthermore, ever since the addition of extra
3688 controls (e.g. 4.43/32), the checks on when to allow different forms of
3689 "control" were broken. There should now be diagnostics for all cases when a
3690 control that does not make sense is encountered.
3692 7. $recipients is now available in the predata ACL (oversight).
3694 8. Tidy the search cache before the fork to do a delivery from a message
3695 received from the command line. Otherwise the child will trigger a lookup
3696 failure and thereby defer the delivery if it tries to use (for example) a
3697 cached ldap connection that the parent has called unbind on.
3699 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3700 of $address_data from the recipient verification was clobbered by the
3701 sender verification.
3703 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3704 was its contents. (It was OK if the option was not defined at all.)
3706 11. A "Completed" log line is now written for messages that are removed from
3707 the spool by the -Mrm option.
3709 12. $host_address is now set to the target address during the checking of
3710 ignore_target_hosts.
3712 13. When checking ignore_target_hosts for an ipliteral router, no host name was
3713 being passed; this would have caused $sender_host_name to have been used if
3714 matching the list had actually called for a host name (not very likely,
3715 since this list is usually IP addresses). A host name is now passed as
3718 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3719 code that specifies a non-restarting handler (typically sigaction() in
3720 modern systems) in an attempt to fix a rare and obscure crash bug.
3722 15. Narrowed the window for a race in the daemon that could cause it to ignore
3723 SIGCHLD signals. This is not a major problem, because they are used only to
3724 wake it up if nothing else does.
3726 16. A malformed maildirsize file could cause Exim to calculate negative values
3727 for the mailbox size or file count. Odd effects could occur as a result.
3728 The maildirsize information is now recalculated if the size or filecount
3731 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3732 support for a long time. Removed HAVE_SYS_VFS_H.
3734 18. Updated exipick to current release from John Jetmore.
3736 19. Allow an empty sender to be matched against a lookup in an address list.
3737 Previously the only cases considered were a regular expression, or an
3740 20. Exim went into a mad DNS lookup loop when doing a callout where the
3741 host was specified on the transport, if the DNS lookup yielded more than
3744 21. The RFC2047 encoding function was originally intended for short strings
3745 such as real names; it was not keeping to the 75-character limit for
3746 encoded words that the RFC imposes. It now respects the limit, and
3747 generates multiple encoded words if necessary. To be on the safe side, I
3748 have increased the buffer size for the ${rfc2047: expansion operator from
3751 22. Failure to deliver a bounce message always caused it to be frozen, even if
3752 there was an errors_to setting on the router. The errors_to setting is now
3755 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3756 canonical form (fully expanded) before being placed in
3757 $sender_host_address.
3759 24. Updated eximstats to version 1.33
3761 25. Include certificate and key file names in error message when GnuTLS fails
3762 to set them up, because the GnuTLS error message doesn't include the name
3763 of the failing file when there is a problem reading it.
3765 26. Expand error message when OpenSSL has problems setting up cert/key files.
3768 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
3769 (this can affect the format of dates).
3771 28. exim_tidydb, when checking for the continued existence of a message for
3772 which it has found a message-specific retry record, was not finding
3773 messages that were in split spool directories. Consequently, it was
3774 deleting retry records that should have stayed in existence.
3776 29. eximstats updated to version 1.35
3777 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
3778 1.35 - bugfix such that pie charts by volume are generated correctly
3780 30. The SPA authentication driver was not abandoning authentication and moving
3781 on to the next authenticator when an expansion was forced to fail,
3782 contradicting the general specification for all authenticators. Instead it
3783 was generating a temporary error. It now behaves as specified.
3785 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
3786 (the order specifies the preference for clients). The order is now AES256,
3787 AES128, 3DES, ARCFOUR128.
3789 31. Small patch to Sieve code - explicitly set From: when generating an
3792 32. Exim crashed if a remote delivery caused a very long error message to be
3793 recorded - for instance if somebody sent an entire SpamAssassin report back
3794 as a large number of 550 error lines. This bug was coincidentally fixed by
3795 increasing the size of one of Exim's internal buffers (big_buffer) that
3796 happened as part of the Exiscan merge. However, to be on the safe side, I
3797 have made the code more robust (and fixed the comments that describe what
3800 33. Some experimental protocols are using DNS PTR records for new purposes. The
3801 keys for these records are domain names, not reversed IP addresses. The
3802 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3803 leaves it alone. Component reversal etc. now happens only for IP addresses.
3806 34. The host_aton() function is supposed to be passed a string that is known
3807 to be a valid IP address. However, in the case of IPv6 addresses, it was
3808 not checking this. This is a hostage to fortune. Exim now panics and dies
3809 if the condition is not met. A case was found where this could be provoked
3810 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3811 components; fortuitously, this particular loophole had already been fixed
3812 by change 4.50/55 or 4.44/33 above.
3814 If there are any other similar loopholes, the new check in host_aton()
3815 itself should stop them being exploited. The report I received stated that
3816 data on the command line could provoke the exploit when Exim was running as
3817 exim, but did not say which command line option was involved. All I could
3818 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3819 running as the user.
3822 35. There was a buffer overflow vulnerability in the SPA authentication code
3823 (which came originally from the Samba project). I have added a test to the
3824 spa_base64_to_bits() function which I hope fixes it.
3827 36. The daemon start-up calls getloadavg() while still root for those OS that
3828 need the first call to be done as root, but it missed one case: when
3829 deliver_queue_load_max is set with deliver_drop_privilege. This is
3830 necessary for the benefit of the queue runner, because there is no re-exec
3831 when deliver_drop_privilege is set.
3833 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3834 was in use, was not putting the data itself into the right store pool;
3835 consequently, it could be overwritten for a subsequent message in the same
3836 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3839 38. Sometimes the final signoff response after QUIT could fail to get
3840 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3841 < 0 before doing a fflush(). This bug looks as though it goes back to the
3842 introduction of TLS in release 3.20, but "sometimes" must have been rare
3843 because the tests only now provoked it.
3849 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
3850 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
3851 Now it returns an integer. A number of calls were still expecting a T/F
3852 return. Fortuitously, in all cases, the tests worked in OK situations,
3853 which is the norm. However, things would have gone wrong on any write
3854 failures on the smtp file descriptor. This function is used when sending
3855 messages over SMTP and also when doing verify callouts.
3857 2. When Exim is called to do synchronous delivery of a locally submitted
3858 message (the -odf or -odi options), it no longer closes stderr before doing
3861 3. Implemented the mua_wrapper option.
3863 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
3865 5. Implemented the functions header_remove(), header_testname(),
3866 header_add_at_position(), and receive_remove_recipient(), and exported them
3869 6. If an ACL "warn" statement specified the addition of headers, Exim already
3870 inserted X-ACL-Warn: at the start if there was no header name. However, it
3871 was not making this test for the second and subsequent header lines if
3872 there were newlines in the string. This meant that an invalid header could
3873 be inserted if Exim was badly configured.
3875 7. Allow an ACL "warn" statement to add header lines at the start or after all
3876 the Received: headers, as well as at the end.
3878 8. Added the rcpt_4xx retry error code.
3880 9. Added postmaster_mailfrom=xxx to callout verification option.
3882 10. Added mailfrom=xxxx to the callout verification option, for verify=
3885 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
3886 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
3888 12. Inserted some casts to stop certain compilers warning when using pointer
3889 differences as field lengths or precisions in printf-type calls (mostly
3890 affecting debugging statements).
3892 13. Added optional readline() support for -be (dynamically loaded).
3894 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
3895 same clock tick as a message's arrival, so that its received time was the
3896 same as the "first fail" time on the retry record, and that message
3897 remained on the queue past the ultimate address timeout, every queue runner
3898 would try a delivery (because it was past the ultimate address timeout) but
3899 after another failure, the ultimate address timeout, which should have then
3900 bounced the address, did not kick in. This was a "< instead of <=" error;
3901 in most cases the first failure would have been in the next clock tick
3902 after the received time, and all would be well.
3904 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
3905 being recognized when the domain list was tested by the match_domain
3906 condition in an expansion string.
3908 16. Added the ${str2b64: operator.
3910 17. Exim was always calling setrlimit() to set a large limit for the number of
3911 processes, without checking whether the existing limit was already
3912 adequate. (It did check for the limit on file descriptors.) Furthermore,
3913 errors from getrlimit() and setrlimit() were being ignored. Now they are
3914 logged to the main and panic logs, but Exim does carry on, to try to do its
3915 job under whatever limits there are.
3917 18. Imported PCRE 5.0.
3919 19. Trivial typo in log message " temporarily refused connection" (the leading
3922 20. If the log selector return_path_on_delivery was set and an address was
3923 redirected to /dev/null, the delivery process crashed because it assumed
3924 that a return path would always be set for a "successful" delivery. In this
3925 case, the whole delivery is bypassed as an optimization, and therefore no
3928 21. Internal re-arrangement: the function for sending a challenge and reading
3929 a response while authentication was assuming a zero-terminated challenge
3930 string. It's now changed to take a pointer and a length, to allow for
3931 binary data in such strings.
3933 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3935 23. Exim was not respecting finduser_retries when seeking the login of the
3936 uid under which it was called; it was always trying 10 times. (The default
3937 setting of finduser_retries is zero.) Also, it was sleeping after the final
3938 failure, which is pointless.
3940 24. Implemented tls_on_connect_ports.
3942 25. Implemented acl_smtp_predata.
3944 26. If the domain in control=submission is set empty, Exim assumes that the
3945 authenticated id is a complete email address when it generates From: or
3946 Sender: header lines.
3948 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3949 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3950 chown and chgrp in /bin and hostname in /usr/bin.
3952 28. Exim was keeping the "process log" file open after each use, just as it
3953 does for the main log. This opens the possibility of it remaining open for
3954 long periods when the USR1 signal hits a daemon. Occasional processlog
3955 errors were reported, that could have been caused by this. Anyway, it seems
3956 much more sensible not to leave this file open at all, so that is what now
3959 29. The long-running daemon process does not normally write to the log once it
3960 has entered its main loop, and it closes the log before doing so. This is
3961 so that log files can straightforwardly be renamed and moved. However,
3962 there are a couple of unusual error situations where the daemon does write
3963 log entries, and I had neglected to close the log afterwards.
3965 30. The text of an SMTP error response that was received during a remote
3966 delivery was being truncated at 512 bytes. This is too short for some of
3967 the long messages that one sometimes sees. I've increased the limit to
3970 31. It is now possible to make retry rules that apply only when a message has a
3971 specific sender, in particular, an empty sender.
3973 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3974 it possible to be selective about when SMTP synchronization is enforced.
3976 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3978 32. Implemented hosts_connection_nolog.
3980 33. Added an ACL for QUIT.
3982 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3985 35. Added mailbox_size and mailbox_filecount to appendfile.
3987 36. Added control = no_multiline_responses to ACLs.
3989 37. There was a bug in the logic of the code that waits for the clock to tick
3990 in the case where the clock went backwards by a substantial amount such
3991 that the microsecond fraction of "now" was more than the microsecond
3992 fraction of "then" (but the whole seconds number was less).
3994 38. Added support for the libradius Radius client library this is found on
3995 FreeBSD (previously only the radiusclient library was supported).
4001 1. When certain lookups returned multiple values in the form name=value, the
4002 quoting of the values was not always being done properly. Specifically:
4003 (a) If the value started with a double quote, but contained no whitespace,
4005 (b) If the value contained whitespace other than a space character (i.e.
4006 tabs or newlines or carriage returns) it was not quoted.
4007 This fix has been applied to the mysql and pgsql lookups by writing a
4008 separate quoting function and calling it from the lookup code. The fix
4009 should probably also be applied to nisplus, ibase and oracle lookups, but
4010 since I cannot test any of those, I have not disturbed their existing code.
4012 2. A hit in the callout cache for a specific address caused a log line with no
4013 reason for rejecting RCPT. Now it says "Previous (cached) callout
4014 verification failure".
4016 3. There was an off-by-one bug in the queryprogram router. An over-long
4017 return line was truncated at 256 instead of 255 characters, thereby
4018 overflowing its buffer with the terminating zero. As well as fixing this, I
4019 have increased the buffer size to 1024 (and made a note to document this).
4021 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4022 when Exim is waiting for an SMTP response from a remote server, Exim
4023 restarts its select() call on the socket, thereby resetting its timeout.
4024 This is not a problem when such interrupts are rare. Somebody set up a cron
4025 job to run exiwhat every 2 minutes, which is less than the normal select()
4026 timeout (5 or 10 minutes). This meant that the select() timeout never
4027 kicked in because it was always reset. I have fixed this by comparing the
4028 time when an interrupt arrives with the time at the start of the first call
4029 to select(). If more time than the timeout has elapsed, the interrupt is
4030 treated as a timeout.
4032 5. Some internal re-factoring in preparation for the addition of Sieve
4033 extensions (by MH). In particular, the "personal" test is moved to a
4034 separate function, and given an option for scanning Cc: and Bcc: (which is
4035 not set for Exim filters).
4037 6. When Exim created an email address using the login of the caller as the
4038 local part (e.g. when creating a From: or Sender: header line), it was not
4039 quoting the local part when it contained special characters such as @.
4041 7. Installed new OpenBSD configuration files.
4043 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4044 try to make them clearer.
4046 9. Callout options, other than the timeout value, were being ignored when
4047 verifying sender addresses in header lines. For example, when using
4049 verify = header_sender/callout=no_cache
4051 the cache was (incorrectly) being used.
4053 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4054 only the Cygwin environment.
4056 11. When return_path_on_delivery was set as a log selector, if different remote
4057 addresses in the same message used different return paths and parallel
4058 remote delivery occurred, the wrong values would sometimes be logged.
4059 (Whenever a remote delivery process finished, the return path value from
4060 the most recently started remote delivery process was logged.)
4062 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4063 lines when AUTH and/or TLS are in use. This is the "received protocol"
4064 field. Exim used to use "asmtp" for authenticated SMTP, without any
4065 indication (in the protocol name) for TLS use. Now it follows the RFC and
4066 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4067 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4068 names appear in log lines as well as in Received: header lines.
4070 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4071 extensions, and comparison tests, and to fix some bugs.
4073 14. Changes to the "personal" filter test:
4075 (1) The test was buggy in that it was just doing the equivalent of
4076 "contains" tests on header lines. For example, if a user's address was
4077 anne@some.where, the "personal" test would incorrectly be true for
4079 To: susanne@some.where
4081 This test is now done by extracting each address from the header in turn,
4082 and checking the entire address. Other tests that are part of "personal"
4083 are now done using regular expressions (for example, to check local parts
4084 of addresses in From: header lines).
4086 (2) The list of non-personal local parts in From: addresses has been
4087 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4088 taken from the Sieve specification recommendations.
4090 (3) If the message contains any header line starting with "List-" it is
4091 treated as non-personal.
4093 (4) The test for "circular" in the Subject: header line has been removed
4094 because it now seems ill-conceived.
4096 15. Minor typos in src/EDITME comments corrected.
4098 16. Installed latest exipick from John Jetmore.
4100 17. If headers_add on a router specified a text string that was too long for
4101 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4102 of string_sprintf() is now avoided.
4104 18. $message_body_size was not set (it was always zero) when running the DATA
4105 ACL and the local_scan() function.
4107 19. For the "mail" command in an Exim filter, no default was being set for
4108 the once_repeat time, causing a random time value to be used if "once" was
4109 specified. (If the value happened to be <= 0, no repeat happened.) The
4110 default is now 0s, meaning "never repeat". The "vacation" command was OK
4111 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4112 (I found it when inspecting the code).
4114 20. There is now an overall timeout for performing a callout verification. It
4115 defaults to 4 times the callout timeout, which applies to individual SMTP
4116 commands during the callout. The overall timeout applies when there is more
4117 than one host that can be tried. The timeout is checked before trying the
4118 next host. This prevents very long delays if there are a large number of
4119 hosts and all are timing out (e.g. when the network connections are timing
4120 out). The value of the overall timeout can be changed by specifying an
4121 additional sub-option for "callout", called "maxwait". For example:
4123 verify = sender/callout=5s,maxwait=20s
4125 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4126 to the end before writing, but this should make it even safer).
4128 22. Exim was forgetting that it had advertised PIPELINING for the second and
4129 subsequent messages on an SMTP connection. It was also not resetting its
4130 memory on STARTTLS and an internal HELO.
4132 23. When Exim logs an SMTP synchronization error within a session, it now
4133 records whether PIPELINING has been advertised or not.
4135 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4136 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4137 rather than long int.
4139 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4141 26. Added the never_mail option to autoreply.
4147 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4148 crash if the getsockname() call failed; this can happen if a connection is
4149 closed very soon after it is established. The problem was simply in the
4150 order in which certain operations were done, causing Exim to try to write
4151 to the SMTP stream before it had set up the file descriptor. The bug has
4152 been fixed by making things happen in the correct order.
4158 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4159 before the connection was closed, thus losing the rejection response.
4161 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4162 some early Solaris releases, but causes trouble in current releases where
4163 socklen_t is defined.
4165 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4168 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4171 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4172 that releasing the top of it at the end releases what was used for sub-
4173 expansions (unless the block got too big). However, discard this block if
4174 the first thing is a variable or header, so that we can use its block when
4175 it is dynamic (useful for very large $message_headers, for example).
4177 6. Lookups now cache *every* query, not just the most recent. A new, separate
4178 store pool is used for this. It can be recovered when all lookup caches are
4179 flushed. Lookups now release memory at the end of their result strings.
4180 This has involved some general refactoring of the lookup sources.
4182 7. Some code has been added to the store_xxx() functions to reduce the amount
4183 of flapping under certain conditions.
4185 8. log_incoming_interface used to affect only the <= reception log lines. Now
4186 it causes the local interface and port to be added to several more SMTP log
4187 lines, for example "SMTP connection from", and rejection lines.
4189 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4191 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4193 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4194 could be overwritten at the end of the current message (or the start of a
4195 new message if it was set in a HELO ACL). The value is now preserved for
4196 the duration of the SMTP connection.
4198 12. If a transport had a headers_rewrite setting, and a matching header line
4199 contained an unqualified address, that address was qualified, even if it
4200 did not match any rewriting rules. The underlying bug was that the values
4201 of the flags that permit the existence of unqualified sender and recipient
4202 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4203 non-local messages, and by -bnq for local messages) were not being
4204 preserved with the message after it was received.
4206 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4207 "next input=" as part of the text comprising the host identity instead of
4208 the correct text. The code was using the same buffer for two different
4209 strings. However, depending on which order the printing function evaluated
4210 its arguments, the bug did not always show up. Under Linux, for example, my
4211 test suite worked just fine.
4213 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4214 doesn't work with some older versions of Perl. It has been changed to "my",
4215 which in any case is probably the better facility to use.
4217 15. A really picky compiler found some instances of statements for creating
4218 error messages that either had too many or two few arguments for the format
4221 16. The size of the buffer for calls to the DNS resolver has been increased
4222 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4223 for addresses that have a lot of PTR records. This alleviates a problem; it
4224 does not fully solve it.
4226 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4227 buffer now truncates the list and logs the incident, which is the same
4228 action as happens when Exim is looking up a host name and its aliases.
4229 Previously in this situation something unpredictable would happen;
4230 sometimes it was "internal error: store_reset failed".
4232 18. If a server dropped the connection unexpectedly when an Exim client was
4233 using GnuTLS and trying to read a response, the client delivery process
4234 crashed while trying to generate an error log message.
4236 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4237 string, for example:
4239 warn message = H1: something\nH2: something
4241 the text was added as a single header line from Exim's point of view
4242 though it ended up OK in the delivered message. However, searching for the
4243 second and subsequent header lines using $h_h2: did not work. This has been
4244 fixed. Similarly, if a system filter added multiple headers in this way,
4245 the routers could not see them.
4247 20. Expanded the error message when iplsearch is called with an invalid key to
4248 suggest using net-iplsearch in a host list.
4250 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4251 ACLs are no longer actually imposed (and a message to that effect is
4254 22. If a "gecos" field in a passwd entry contained escaped characters, in
4255 particular, if it contained a \" sequence, Exim got it wrong when building
4256 a From: or a Sender: header from that name. A second bug also caused
4257 incorrect handling when an unquoted " was present following a character
4258 that needed quoting.
4260 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4261 was not being matched caselessly.
4263 24. Arranged for all hyphens in the exim.8 source to be escaped with
4266 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4267 properly. Recipient callout cache records were still being keyed to include
4268 the sender, even when use_sender was set false. This led to far more
4269 callouts that were necessary. The sender is no longer included in the key
4270 when use_sender is false.
4272 26. Added "control = submission" modifier to ACLs.
4274 27. Added the ${base62d: operator to decode base 62 numbers.
4276 28. dnsdb lookups can now access SRV records.
4278 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4279 the configuration file.
4281 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4282 (-v) mode. This makes the output for a verbose queue run more intelligible.
4284 31. Added a use_postmaster feature to recipient callouts.
4286 32. Added the $body_zerocount variable, containing the number of binary zero
4287 bytes in the message body.
4289 33. The time of last modification of the "new" subdirectory is now used as the
4290 "mailbox time last read" when there is a quota error for a maildir
4293 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4295 35. Added +ignore_unknown as a special item in host lists.
4297 36. Code for decoding IPv6 addresses in host lists is now included, even if
4298 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4299 address was recognized as an IP address, but was then not correctly decoded
4300 into binary, causing unexpected and incorrect effects when compared with
4307 1. Very minor rewording of debugging text in manualroute to say "list of
4308 hosts" instead of "hostlist".
4310 2. If verify=header_syntax was set, and a header line with an unqualified
4311 address (no domain) and a large number of spaces between the end of the
4312 name and the colon was received, the reception process suffered a buffer
4313 overflow, and (when I tested it) crashed. This was caused by some obsolete
4314 code that should have been removed. The fix is to remove it!
4316 3. When running in the test harness, delay a bit after writing a bounce
4317 message to get a bit more predictability in the log output.
4319 4. Added a call to search_tidyup() just before forking a reception process. In
4320 theory, someone could use a lookup in the expansion of smtp_accept_max_
4321 per_host which, without the tidyup, could leave open a database connection.
4323 5. Added the variables $recipient_data and $sender_data which get set from a
4324 lookup success in an ACL "recipients" or "senders" condition, or a router
4325 "senders" option, similar to $domain_data and $local_part_data.
4327 6. Moved the writing of debug_print from before to after the "senders" test
4330 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4331 problems for message scanning, either using a data ACL, or using
4332 local_scan() because the Received: header was not generated till after they
4333 were called (in order to set the time as the time of reception completion).
4334 I have revised the way this works. The header is now generated after the
4335 body is received, but before the ACL or local_scan() are called. After they
4336 are run, the timestamp in the header is updated.
4342 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4343 before starting a queue runner without re-exec. This happened only when
4344 deliver_drop_privilege was set or when the Exim user was set to root. The
4345 effect of the bug was that timeouts during subsequent deliveries caused
4346 crashes instead of being properly handled. The handler is now left at its
4347 default (and expected) setting.
4349 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4350 message, again when deliver_drop_privilege is set or Exim is run as root.
4351 The bug described in (1) was not present in this case, but the tidying up
4352 of the other signals was missing. I have made the two cases consistent.
4354 3. The ignore_target_hosts setting on a manualroute router was being ignored
4355 for hosts that were looked up using the /MX notation.
4357 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4360 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4361 operated on the sender address. After changing the $sender_address to <>
4362 for the sender address verify, Exim was re-instated it as the original
4363 (before rewriting) address, but remembering that it had rewritten it, so it
4364 wasn't rewriting it again. This bug also had the effect of breaking the
4365 sender address verification caching when the sender address was rewritten.
4367 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4368 This has been changed so that if the ip literal address matches
4369 ignore_target_hosts, the router declines.
4371 7. Added expansion conditions match_domain, match_address, and match_local_
4372 part (NOT match_host).
4374 8. The placeholder for the Received: header didn't have a length field set.
4376 9. Added code to Exim itself and to exim_lock to test for a specific race
4377 condition that could lead to file corruption when using MBX delivery. The
4378 issue is with the lockfile that is created in /tmp. If this file is removed
4379 after a process has opened it but before that process has acquired a lock,
4380 there is the potential for a second process to recreate the file and also
4381 acquire a lock. This could lead to two Exim processes writing to the file
4382 at the same time. The added code performs the same test as UW imapd; it
4383 checks after acquiring the lock that its file descriptor still refers to
4384 the same named file.
4386 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4387 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4390 11. Added the smtp_active_hostname option. If used, this will typically be made
4391 to depend on the incoming interface address. Because $interface_address is
4392 not set up until the daemon has forked a reception process, error responses
4393 that can happen earlier (such as "too many connections") no longer contain
4396 12. If an expansion in a condition on a "warn" statement fails because a lookup
4397 defers, the "warn" statement is abandoned, and the next ACL statement is
4398 processed. Previously this caused the whole ACL to be aborted.
4400 13. Added the iplsearch lookup type.
4402 14. Added ident_timeout as a log selector.
4404 15. Added tls_certificate_verified as a log selector.
4406 16. Added a global option tls_require_ciphers (compare the smtp transport
4407 option of the same name). This controls incoming TLS connections.
4409 17. I finally figured out how to make tls_require_ciphers do a similar thing
4410 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
4411 before starting the TLS session.
4413 18. Tabs are now shown as \t in -bP output.
4415 19. If the log selector return_path_on_delivery was set, Exim crashed when
4416 bouncing a message because it had too many Received: header lines.
4418 20. If two routers both had headers_remove settings, and the first one included
4419 a superfluous trailing colon, the final name in the first list and the
4420 first name in the second list were incorrectly joined into one item (with a
4421 colon in the middle).
4427 1. Added -C and -D options to the exinext utility, mainly to make it easier
4428 to include in the automated testing, but these could be helpful when
4429 multiple configurations are in use.
4431 2. The exinext utility was not formatting the output nicely when there was
4432 an alternate port involved in the retry record key, nor when there was a
4433 message id as well (for retries that were specific to a specific message
4434 and a specific host). It was also confused by IPv6 addresses, because of
4435 the additional colons they contain. I have fixed the IPv4 problem, and
4436 patched it up to do a reasonable job for IPv6.
4438 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
4439 delivery, the log line now contains "pipelined" if PIPELINING was used.
4441 4. An SMTP transport process used to panic and die if the bind() call to set
4442 an explicit outgoing interface failed. This has been changed; it is now
4443 treated in the same way as a connect() failure.
4445 5. A reference to $sender_host_name in the part of a conditional expansion
4446 that was being skipped was still causing a DNS lookup. This no longer
4449 6. The def: expansion condition was not recognizing references to header lines
4450 that used bh_ and bheader_.
4452 7. Added the _cache feature to named lists.
4454 8. The code for checking quota_filecount in the appendfile transport was
4455 allowing one more file than it should have been.
4457 9. For compatibility with Sendmail, the command line option
4465 and sets the incoming protocol and host name (for trusted callers). The
4466 host name and its colon can be omitted when only the protocol is to be set.
4467 Note the Exim already has two private options, -pd and -ps, that refer to
4468 embedded Perl. It is therefore impossible to set a protocol value of "d" or
4469 "s", but I don't think that's a major issue.
4471 10. A number of refactoring changes to the code, none of which should affect
4474 (a) The number of logging options was getting close to filling up the
4475 32-bit word that was used as a bit map. I have split them into two classes:
4476 those that are passed in the argument to log_write(), and those that are
4477 only ever tested independently outside of that function. These are now in
4478 separate 32-bit words, so there is plenty of room for expansion again.
4479 There is no change in the user interface or the logging behaviour.
4481 (b) When building, for example, log lines, the code previously used a
4482 macro that called string_cat() twice, in order to add two strings. This is
4483 not really sufficiently general. Furthermore, there was one instance where
4484 it was actually wrong because one of the argument was used twice, and in
4485 one call a function was used. (As it happened, calling the function twice
4486 did not affect the overall behaviour.) The macro has been replaced by a
4487 function that can join an arbitrary number of extra strings onto a growing
4490 (c) The code for expansion conditions now uses a table and a binary chop
4491 instead of a serial search (which was left over from when there were very
4492 few conditions). Also, it now recognizes conditions like "pam" even when
4493 the relevant support is not compiled in: a suitably worded error message is
4494 given if an attempt is made to use such a condition.
4496 11. Added ${time_interval:xxxxx}.
4498 12. A bug was causing one of the ddress fields not to be passed back correctly
4499 from remote delivery subprocesses. The field in question was not being
4500 subsequently used, so this caused to problems in practice.
4502 13. Added new log selectors queue_time and deliver_time.
4504 14. Might have fixed a bug in maildirsizefile handling that threw up
4505 "unexpected character" debug warnings, and recalculated the data
4506 unnecessarily. In any case, I expanded the warning message to give more
4509 15. Added the message "Restricted characters in address" to the statements in
4510 the default ACL that block characters like @ and % in local parts.
4512 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
4513 Three changes have been made:
4515 (a) There was a serious bug; a negative response to MAIL caused the whole
4516 recipient domain to be cached as invalid, thereby blocking all messages
4517 to all local parts at the same domain, from all senders. This bug has
4518 been fixed. The domain is no longer cached after a negative response to
4519 MAIL if the sender used is not empty.
4521 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
4524 (c) A new callout option, "use_sender" has been added for people who want
4525 the modified behaviour.
4531 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
4534 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
4535 indeed breaks things for older releases.
4537 3. Added additional logging to the case where there is a problem reading data
4538 from a filter that is running in a subprocess using a pipe, in order to
4539 try to track down a specific problem.
4541 4. Testing facility fudge: when running in the test harness and attempting
4542 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
4543 getting "No route to host". Convert this to a timeout.
4545 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
4548 6. Some OS don't have socklen_t but use size_t instead. This affects the
4549 fifth argument of getsockopt() amongst other things. This is now
4550 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
4551 can be set for individual OS. I have set it for SunOS5, OSF1, and
4552 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
4553 some earlier ones do not.
4555 7. Change 4.30/15 was not doing the test caselessly.
4557 8. The standard form for an IPv6 address literal was being rejected by address
4558 parsing in, for example, MAIL and RCPT commands. An example of this kind of
4559 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4560 this, as well as the form without the "IPv6" on the front (but only when
4561 address literals are enabled, of course).
4563 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
4565 10. Exim crashed if a message with an empty sender address specified by -f
4566 encountered a router with an errors_to setting. This could be provoked only
4567 by a command such as
4571 where an empty string was supplied; "<>" did not hit this bug.
4573 11. Installed PCRE release 4.5.
4575 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
4576 remained set. It is now erased.
4578 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
4579 times from message ids (which are base 36 rather than the normal 62).
4581 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
4582 were being counted as actual protocol errors, and logged if the log
4583 selector +smtp_protocol_error was set. One cannot be perfect in this test,
4584 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
4585 and DATA following a set of rejected RCPTs do not count as protocol errors.
4586 In other words, Exim assumes they were pipelined, though this may not
4587 actually be the case. Of course, in all cases the client gets an
4588 appropriate error code.
4590 15. If a lookup fails in an ACL condition, a message about the failure may
4591 be available; it is used if testing the ACL cannot continue, because most
4592 such messages specify what the cause of the deferral is. However, some
4593 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
4594 that caused an old message to be retained and used if a later statement
4595 caused a defer, replacing the real cause of the deferral.
4597 16. If an IP address had so many PTR records that the DNS lookup buffer
4598 was not large enough to hold them, Exim could crash while trying to process
4599 the truncated data. It now detects and logs this case.
4601 17. Further to 4.21/58, another change has been made: if (and only if) the
4602 first line of a message (the first header line) ends with CRLF, a bare LF
4603 in a subsequent header line has a space inserted after it, so as not to
4604 terminate the header.
4606 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
4607 unnecessarily, used atoi() instead of strtol(), and didn't check the
4608 termination when getting file sizes from file names by regex.
4610 19. Completely re-implemented the support for maildirsize files, in the light
4611 of a number of problems with the previous contributed implementation
4612 (4.30/29). In particular:
4614 . If the quota is zero, the maildirsize file is maintained, but no quota is
4617 . If the maildir directory does not exist, it is created before any attempt
4618 to write a maildirsize file.
4620 . The quota value in the file is just a cache; if the quota is changed in
4621 the transport, the new value overrides.
4623 . A regular expression is available for excluding directories from the
4626 20. The autoreply transport checks the characters in options that define the
4627 message's headers; it allows continued headers, but it was checking with
4628 isspace() after an embedded newline instead of explicitly looking for a
4631 21. If all the "regular" hosts to which an address was routed had passed their
4632 expiry times, and had not reached their retry times, the address was
4633 bounced, even if fallback hosts were defined. Now Exim should go on to try
4636 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
4637 equivalent code in the SMTP transport. Some hosts send humungous responses
4638 to HELO/EHLO, more than 1024 it seems.
4640 23. Refactoring: code in filter.c used (void *) for "any old type" but this
4641 gives compiler warnings in some environments. I've now done it "properly",
4644 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
4645 (because of problems with the built-in one) was declared to return uschar *
4646 instead of char *, causing compiler failure.
4648 25. Fixed a file descriptor leak when processing alias/forward files.
4650 26. Fixed a minor format string issue in dbfn.c.
4652 27. Typo in exim.c: ("dmbnz" for "dbmnz").
4654 28. If a filter file refered to $h_xxx or $message_headers, and the headers
4655 contained RFC 2047 "words", Exim's memory could, under certain conditions,
4658 29. When a sender address is verified, it is cached, to save repeating the test
4659 when there is more than one recipient in a message. However, when the
4660 verification involves a callout, it is possible for different callout
4661 options to be set for different recipients. It is too complicated to keep
4662 track of this in the cache, so now Exim always runs a verification when a
4663 callout is required, relying on the callout cache for the optimization.
4664 The overhead is duplication of the address routing, but this should not be
4667 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
4668 to be verified with callout=postmaster, and the main callout worked but the
4669 postmaster check failed, the verification correctly failed. However, if a
4670 subsequent RCPT command asked for sender verification *without* the
4671 postmaster check, incorrect caching caused this verification also to fail,
4674 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
4675 it was not caching the DNS options (qualify_single, search_parents) that
4676 were used when the lookup failed. A subsequent lookup with different
4677 options therefore always gave the same answer, though there were cases
4678 where it should not have. (Example: a "domains = !$mx_any" option on a
4679 dnslookup router: the "domains" option is always processed without any
4680 widening, but the router might have qualify_single set.) Now Exim uses the
4681 cached value only when the same options are set.
4683 32. Added John Jetmore's "exipick" utility to the distribution.
4685 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
4686 than a timeout (e.g. a certificate is required, and is not provided), an
4687 Exim server now closes the connection immediately. Previously it waited for
4688 the client to close - but if the client is SSL, it seems that they each
4689 wait for each other, leading to a delay before one of them times out.
4691 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
4692 maintained 0.8.x compatibility because I don't think many are using it, and
4693 it is clearly obsolete.
4695 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
4698 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
4699 client certificate was expired. A simple patch fixes this, though I don't
4700 understand the full logic of why the verify callback is called multiple
4703 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
4704 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
4705 which causes problems with some clients (such as the Certicom SSL Plus
4706 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
4707 disables the coutermeasure allowing Eudora to connect."
4709 38. Exim was not checking that a write() to a log file succeeded. This could
4710 lead to Bad Things if a log got too big, in particular if it hit a file
4711 size limit. Exim now panics and dies if it cannot write to a log file, just
4712 as it does if it cannot open a log file.
4714 39. Modified OS/Makefile-Linux so that it now contains
4716 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
4718 The two -D definitions ensure that Exim is compiled with large file
4719 support, which makes it possible to handle log files that are bigger than
4722 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
4723 instance) a domain was checked against a named list that involved a lookup,
4724 causing $domain_data to be set, then another domain was checked against the
4725 same list, then the first domain was re-checked, the value of $domain_data
4726 after the final check could be wrong. In particular, if the second check
4727 failed, it could be set empty. This bug probably also applied to
4730 41. The strip_trailing_dot option was not being applied to the address given
4731 with the -f command-line option.
4733 42. The code for reading a message's header from the spool was incrementing
4734 $received_count, but never initializing it. This meant that the value was
4735 incorrect (doubled) while delivering a message in the same process in which
4736 it was received. In the most common configuration of Exim, this never
4737 happens - a fresh exec is done - but it can happen when
4738 deliver_drop_privilege is set.
4740 43. When Exim logs an SMTP synchronization error - client data sent too soon -
4741 it now includes up to 150 characters of the unexpected data in the log
4744 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
4745 and building data strings. The size of both of these buffers was 10 000
4746 bytes - far larger than anybody would *ever* want, thought I. Needless to
4747 say, somebody hit the limit. I have increased the maximum line length to
4748 20 000 and the maximum data length of concatenated lines to 100 000. I have
4749 also fixed two bugs, because there was no checking on these buffers. Tsk,
4750 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
4751 buffer is too small.
4753 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
4754 lsearch lookups. Now it does.
4756 46. When parsing a route_list item in a manualroute router, a fixed-length
4757 buffer was used for the list of hosts. I made this 1024 bytes long,
4758 thinking that nobody would ever have a list of hosts that long. Wrong.
4759 Somebody had a whole pile of complicated expansion conditions, and the
4760 string was silently truncated, leading to an expansion error. It turns out
4761 that it is easier to change to an unlimited length (owing to other changes
4762 that have happened since this code was originally written) than to build
4763 structure for giving a limitation error. The length of the item that
4764 expands into the list of hosts is now unlimited.
4766 47. The lsearch lookup could not handle data where the length of text line was
4767 more than 4095 characters. Such lines were truncated, leading to shortened
4768 data being returned. It should now handle lines of any length.
4770 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
4771 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
4774 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
4775 replacing strings such as BIN_DIRECTORY in a source file: the replacement
4776 no longer happens in comment lines. A list of replacements is now placed
4777 at the head of all of the source files, except those whose only change is
4778 to replace PERL_COMMAND in the very first #! line.
4780 50. Replaced the slow insertion sort in queue.c, for sorting the list of
4781 messages on the queue, with a bottom-up merge sort, using code contributed
4782 by Michael Haardt. This should make operations like -bp somewhat faster on
4783 large queues. It won't affect queue runners, except when queue_run_in_order
4786 51. Installed eximstats 1.31 in the distribution.
4788 52. Added support for SRV lookups to the dnslookup router.
4790 53. If an ACL referred to $message_body or $message_body_end, the value was not
4791 reset for any messages that followed in the same SMTP session.
4793 54. The store-handling optimization for building very long strings was not
4794 differentiating between the different store pools. I don't think this
4795 actually made any difference in practice, but I've tidied it.
4797 55. While running the routers to verify a sender address, $sender_address
4798 was still set to the sender address. This is wrong, because when routing to
4799 send a bounce to the sender, it would be empty. Therefore, I have changed
4800 it so that, while verifying a sender address, $sender_address is set to <>.
4801 (There is no change to what happens when verifying a recipient address.)
4803 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
4804 target A or AAAA records (if not already returned) without resetting the
4805 qualify_single or search_parents options of the DNS resolver. These are
4806 inappropriate in this case because the targets of MX and SRV records must
4807 be FQDNs. A broken DNS record could cause trouble if it happened to have a
4808 target that, when qualified, matched something in the local domain. These
4809 two options are now turned off when doing these lookups.
4811 57. It seems that at least some releases of Reiserfs (which does not have the
4812 concept of a fixed number of inodes) returns zero and not -1 for the
4813 number of available inodes. This interacted badly with check_spool_inodes,
4814 which assumed that -1 was the "no such thing" setting. What I have done is
4815 to check that the total number of inodes is greater than zero before doing
4816 the test of how many are available.
4818 58. When a "warn" ACL statement has a log_message modifier, the message is
4819 remembered, and not repeated. This is to avoid a lot of repetition when a
4820 message has many recipients that cause the same warning to be written.
4821 Howewer, Exim was preserving the list of already written lines for an
4822 entire SMTP session, which doesn't seem right. The memory is now reset if a
4823 new message is started.
4825 59. The "rewrite" debugging flag was not showing the result of rewriting in the
4826 debugging output unless log_rewrite was also set.
4828 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
4829 of (int)(handle) when we know that handle contains (void *)(-1).
4831 61. The Exim daemon panic-logs an error return when it closes the incoming
4832 connection. However "connection reset by peer" seems to be common, and
4833 isn't really an error worthy of noting specially, so that particular error
4836 62. When Exim is trying to find all the local interfaces, it used to panic and
4837 die if the ioctl to get the interface flags failed. However, it seems that
4838 on at least one OS (Solaris 9) it is possible to have an interface that is
4839 included in the list of interfaces, but for which you get a failure error
4840 for this call. This happens when the interface is not "plumbed" into a
4841 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
4842 failure of the "get flags" call assumes that the interface is down.
4844 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
4845 makes life easier for people who are doing arithmetic on fields extracted
4846 from dates, where you often get leading zeros that should not be
4847 interpreted as octal.
4849 64. Added qualify_domain to the redirect router, to override the global
4852 65. If a pathologically long header line contained very many addresses (the
4853 report of this problem mentioned 10 000) and each of them was rewritten,
4854 Exim could use up a very large amount of memory. (It kept on making new
4855 copies of the header line as it rewrote, and never released the old ones.)
4856 At the expense of a bit more processing, the header rewriting function has
4857 been changed so that it no longer eats memory in this way.
4859 66. The generation of the Received: header has been moved from the time that a
4860 message starts to be received, to the time that it finishes. The timestamp
4861 in the Received: header should now be very close to that of the <= log
4862 line. There are two side-effects of this change:
4864 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
4865 logged header lines no longer include the local Received: line, because
4866 it has not yet been created. The same applies to a copy of the message
4867 that is returned to a non-SMTP sender when a message is rejected.
4869 (b) When a filter file is tested using -bf, no additional Received: header
4870 is added to the test message. After some thought, I decided that this
4873 This change does not affect the value of $received_for. It is still set
4874 after address rewriting, but before local_scan() is called.
4876 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
4878 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
4879 gave an unhelpful panic error message, and a defer error. I have managed to
4880 change this behaviour so that it now rejects any supplied certificate,
4881 which seems right, as the list of acceptable certificates is empty.
4883 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
4884 gave an unhelpful defer error. I have not managed to make this reject any
4885 supplied certificates, but the error message it gives is "no certificate
4886 supplied", which is not helpful.
4888 70. exigrep's output now also includes lines that are not associated with any
4889 message, but which match the given pattern. Implemented by a patch from
4890 Martin Sluka, which also tidied up the Perl a bit.
4892 71. Recipient callout verification, like sender verification, was using <> in
4893 the MAIL FROM command. This isn't really the right thing, since the actual
4894 sender may affect whether the remote host accepts the recipient or not. I
4895 have changed it to use the actual sender in the callout; this means that
4896 the cache record is now keyed on a recipient/sender pair, not just the
4897 recipient address. There doesn't seem to be a real danger of callout loops,
4898 since a callout by the remote host to check the sender would use <>.
4899 [SEE ABOVE: changed after hitting problems.]
4901 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
4902 temporary errors. However, in the case of such a code being given after
4903 the end of a data transmission (i.e. after ".") Exim was failing to write
4904 a retry record for the message. (Yes, there was some broken host that was
4905 actually sending 8xx at this point.)
4907 73. An unknown lookup type in a host list could cause Exim to panic-die when
4908 the list was checked. (An example that provoked this was putting <; in the
4909 middle of a list instead of at the start.) If this happened during a DATA
4910 ACL check, a -D file could be left lying around. This kind of configuration
4911 error no longer causes Exim to die; instead it causes a defer errror. The
4912 incident is still logged to the main and panic logs.
4914 74. Buglet left over from Exim 3 conversion. The message "too many messages
4915 in one connection" was written to the rejectlog but not the mainlog, except
4916 when address rewriting (yes!) was being logged.
4918 75. Added write_rejectlog option.
4920 76. When a system filter was run not as root (that is, when system_filter_user
4921 was set), the values of the $n variables were not being returned to the
4922 main process; thus, they were not subsequently available in the $sn
4925 77. Added +return_path_on_delivery log selector.
4927 78. A connection timeout was being treated differently from recipients deferred
4928 when testing hosts_max_try with a message that was older than the host's
4929 retry timeout. (The host should not be counted, thus allowing all hosts to
4930 be tried at least once before bouncing.) This may have been the cause of an
4931 occasionally reported bug whereby a message would remain on the queue
4932 longer than the retry timeout, but would be bounced if a delivery was
4933 forced. I say "may" because I never totally pinned down the problem;
4934 setting up timeout/retry tests is difficult. See also the next item.
4936 79. The ultimate address timeout was not being applied to errors that involved
4937 a combination of host plus message (for example, a timeout on a MAIL
4938 command). When an address resolved to a number of possible hosts, and they
4939 were not all tried for each delivery (e.g. because of hosts_max_try), a
4940 message could remain on the queue longer than the retry timeout.
4942 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4945 81. Fixed an obscure SMTP outgoing bug which required at least the following
4946 conditions: (a) there was another message waiting for the same server;
4947 (b) the server returned 5xx to all RCPT commands in the first message so
4948 that the message was not completed; (c) the server dropped the connection
4949 or gave a negative response to the RSET that Exim sends to abort the
4950 transaction. The observed case was a dropped connection after DATA that had
4951 been sent in pipelining mode. That is, the server had advertised PIPELINING
4952 but was not implementing it correctly. The effect of the bug was incorrect
4953 behaviour, such as trying another host, and this could lead to a crash.
4959 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4960 and daemon.c were passed as pointers to ints; they should have been
4961 pointers to socklen_t variables (which are typically unsigned ints).
4963 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4966 3. Fixed a really odd bug that affected only the testing scheme; patching a
4967 certain fixed string in the binary changed the value of another string that
4968 happened to be identical to the end of the original first string.
4970 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4971 name", it returns that address as the IP address. On some operating
4972 systems (e.g. Solaris), it also passes back the IP address string as the
4973 "host name". However, on others (e.g. Linux), it passes back an empty
4974 string. Exim wasn't checking for this, and was changing the host name to an
4975 empty string, assuming it had been canonicized.
4977 5. Although rare, it is permitted to have more than one PTR record for a given
4978 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4979 all the names associated with an address, because they do in Solaris.
4980 However, it seems that they do not in Linux for data that comes from the
4981 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4982 I found this out when I moved to a new Linux workstation and tried to run
4983 the Exim test suite.
4985 To get round this problem I have changed the code so that it now does its
4986 own call to the DNS to look up PTR records when searching for a host name.
4987 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
4988 addresses that are only in /etc/hosts are still found.
4990 This behaviour is, however, controlled by an option called host_lookup_
4991 order, which defaults to "bydns:byaddr". If people want to use the other
4992 order, or indeed, just use one or the other means of lookup, they can
4993 specify it in this variable.
4995 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
4996 some operating systems, this comes back from gethostbyaddr() as an empty
4997 string, and this is what Exim used to test for. However, it seems that in
4998 other systems, "." is yielded. Exim now tests for this case too.
5000 7. The values of check_spool_space and check_log_space are now held internally
5001 as a number of kilobytes instead of an absolute number of bytes. If a
5002 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5003 kilobyte. This means that much larger values can be stored.
5005 8. Exim monitor: an attempt to get the action menu when not actually pointing
5006 at a message produces an empty menu entitled "No message selected". This
5007 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5008 no entries in it ("Shell widget menu has zero width and/or height"). So I
5009 have added a single, blank menu entry in this case.
5011 9. Added ${quote_local_part.
5013 10. MIME decoding is now applied to the contents of Subject: header lines when
5016 11. Now that a reference to $sender_host_address automatically causes a reverse
5017 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5018 host lookup before query-style lookups in lists that might use this
5019 variable. This has therefore been abolished, and the "net-" prefix is no
5020 longer necessary for query-style lookups.
5022 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5023 have been a typo for LFLAGS, so it has been changed.
5025 13. The install script calls Exim with "-C /dev/null" in order to find the
5026 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5027 to be output. Howeve, since Exim outputs its version number before the
5028 error, it didn't break the script. It just looked ugly. I fixed this by
5029 always allowing "-C /dev/null" if the caller is root.
5031 14. Ignore overlarge ACL variable number when reading spool file - insurance
5032 against a later release with more variables having written the file.
5034 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5035 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5036 this, as well as the form without the "IPv6" on the front.
5038 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5039 OS/Makefile-Darwin file.
5041 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5042 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5044 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5045 variable. However, some OS use ints for inodes. I've added cast to long int
5046 to get rid of the compiler warning.
5048 19. I had forgotten to lock out "/../" in configuration file names when
5049 ALT_CONFIG_PREFIX was set.
5051 20. Routers used for verification do not need to specify transports. However,
5052 if such a router generated a host list, and callout was configured, Exim
5053 crashed, because it could not find a port number from the (non-existent)
5054 transport. It now assumes port 25 in this circumstance.
5056 21. Added the -t option to exigrep.
5058 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5059 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5060 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5061 them is set, LOOKUP_LSEARCH is forced.
5063 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5065 24. Added sender and host information to the "rejected by local_scan()" log
5066 line; previously there was no indication of these.
5068 25. Added .include_if_exists.
5070 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5071 Linux. It turns out that not all file systems support this. Apparently some
5072 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5073 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5074 not supported on the file descriptor, is now ignored when Exim is trying to
5075 sync a directory. This applies only to Linux.
5077 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5079 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5080 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5083 29. Added support for maildirsize files from supplied patch (modified a bit).
5085 30. The use of :fail: followed by an empty string could lead Exim to respond to
5086 sender verification failures with (e.g.):
5088 550 Verification failed for <xxx>
5089 550 Sender verify failed
5091 where the first response line was missing the '-' that indicates it is not
5092 the final line of the response.
5094 31. The loop for finding the name of the user that called Exim had a hardwired
5095 limit of 10; it now uses the value of finduser_retries, which is used for
5096 all other user lookups.
5098 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5101 33. Exim was neglecting to zero errno before one call of strtol() when
5102 expanding a string and expecting an integer value. On some systems this
5103 resulted in spurious "integer overflow" errors. Also, it was casting the
5104 result into an int without checking.
5106 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5107 did not work. The code looks as if it has *never* worked, though it appears
5108 to have been documented since at least releast 1.62. I have made it work.
5110 35. The "timeout_DNS" error in retry rules, also documented since at least
5111 1.62, also never worked. As it isn't clear exactly what this means, and
5112 clearly it isn't a major issue, I have abolished the feature by treating it
5113 as "timeout", and writing a warning to the main and panic logs.
5115 36. The display of retry rules for -brt wasn't always showing the error code
5118 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5119 timeout_connect_A, timeout_connect_MX.
5121 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5122 to the empty sender.
5124 39. The daemon was not analysing the content of -oX till after it had closed
5125 stderr and disconnected from the controlling terminal. This meant that any
5126 syntax errors were only noted on the panic log, and the return code from
5127 the command was 0. By re-arranging the code a little, I've made the
5128 decoding happen first, so such errors now appear on stderr, and the return
5129 code is 1. However, the actual setting up of the sockets still happens in
5130 the disconnected process, so errors there are still only recorded on the
5133 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5134 connections (as happens on some IP stacks) was logged at start up time as
5135 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5136 it from "IPv6 and IPv4", which means that two separate sockets are being
5139 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5140 says whether AF_INET or AF_INET6 was passed as an argument.
5142 42. Exiwhat output was messed up when time zones were included in log
5145 43. Exiwhat now gives more information about the daemon's listening ports,
5146 and whether -tls-on-connect was used.
5148 44. The "port" option of the smtp transport is now expanded.
5150 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5151 silently ignored. Now an error message is written to the main and panic
5154 46. There's a new ACL modifier called "logwrite" which writes to a log file
5155 as soon as it is encountered.
5157 47. Added $local_user_uid and $local_user_gid at routing time.
5159 48. Exim crashed when trying to verify a sender address that was being
5162 49. Exim was recognizing only a space character after ".include". It now also
5163 recognizes a tab character.
5165 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5166 extracting the relevant information from the specification. The man page no
5167 longer contains scrambled data for the -d option, and I've added a section
5168 at the front about calling Exim under different names.
5170 51. Added "extra_headers" argument to the "mail" command in filter files.
5172 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5175 53. Installed eximstats 1.29.
5177 54. Added transport_filter_timeout as a generic transport option.
5179 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5180 Cc: header lines. This was required by RFC 822, but it not required by RFC
5183 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5184 incoming messages that did not have them. Now it does so only if the
5185 message originates locally, that is, if there is no associated remote host
5186 address. When Resent- header lines are present, this applies to the Resent-
5187 lines rather than the non-Resent- lines.
5189 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5190 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5192 58. Messages for configuration errors now include the name of the main
5193 configuration file - useful now that there may be more than one file in a
5194 list (.included file names were always shown).
5196 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5197 for those rare installations that do not start the daemon as root or run it
5198 setuid root. I've cut out the call to initgroups() if the daemon is not
5201 60. The Exim user and group can now be bound into the binary as text strings
5202 that are looked up at the start of Exim's processing.
5204 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5206 62. Added $mailstore_basename variable.
5208 63. Installed patch to sieve.c from Michael Haardt.
5210 64. When Exim failed to open the panic log after failing to open the main log,
5211 the original message it was trying to log was written to stderr and debug
5212 output, but if they were not available (the usual case in production), it
5213 was lost. Now it is written to syslog before the two lines that record the
5214 failures to open the logs.
5216 65. Users' Exim filters run in subprocesses under the user's uid. It is
5217 possible for a "deliver" command or an alias in a "personal" command to
5218 provoke an address rewrite. If logging of address rewriting is configured,
5219 this fails because the process is not running as root or exim. There may be
5220 a better way of dealing with this, but for the moment (because 4.30 needs
5221 to be released), I have disabled address rewrite logging when running a
5222 filter in a non-root, non-exim process.
5228 1. The buildconfig auxiliary program wasn't quoting the value set for
5229 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5230 not defined. This bug was masked in 4.22 by the effect that was fixed in
5233 2. Some messages that were rejected after a message id was allocated were
5234 shown as "incomplete" by exigrep. It no longer does this for messages that
5235 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5237 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5238 have allow_domain_literals set, the ID did not get logged in the <= line.
5239 Domain literals are now always recognized in Message-ID: header lines.
5241 4. The first argument for a ${extract expansion item is the key name or field
5242 number. Leading and trailing spaces in this item were not being ignored,
5243 causing some misleading effects.
5245 5. When deliver_drop_privilege was set, single queue runner processes started
5246 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5247 same command in the process it spins off) were not dropping privilege.
5249 6. When the daemon running as "exim" started a queue runner, it always
5250 re-executed Exim in the spun-off process. This is a waste of effort when
5251 deliver_drop_privilege is set. The new process now just calls the
5252 queue-runner function directly.
5258 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5261 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5262 ignored. Though the use of -oP was forcing the writing of a pid file, it
5263 was always written to the default place.
5265 3. If the message "no IP address found for host xxxx" is generated during
5266 incoming verification, it is now followed by identification of the incoming
5267 connection (so you can more easily find what provoked it).
5269 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5271 5. Added some features to "harden" Exim a bit more against certain attacks:
5273 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5274 be put in Local/Makefile. This is like the never_users runtime option,
5275 but it cannot be overridden. The default setting is "root".
5277 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5278 prefix string with which any file named in a -C command line option
5281 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5282 is retained for -C and -D only if the caller of Exim is root. Without
5283 it, the exim user may also use -C and -D and retain privilege.
5285 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5286 command line option is disabled.
5288 6. Macro names set by the -D option must start with an upper case letter, just
5289 like macro names defined in the configuration file.
5291 7. Added "dereference=" facility to LDAP.
5293 8. Two instances of the typo "uknown" in the source files are fixed.
5295 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5296 the Configure-Makefile script screwed up while processing it.
5298 10. Incorporated PCRE 4.4.
5300 11. The SMTP synchronization check was not operating right at the start of an
5301 SMTP session. For example, it could not catch a HELO sent before the client
5302 waited for the greeting. There is now a check for outstanding input at the
5303 point when the greeting is written. Because of the duplex, asynchronous
5304 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5305 way, but not yet received, when the check is performed.
5307 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5308 on TCP/IP sockets, because this apparently causes some broken clients to
5311 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5312 unchanged) from the Cygwin maintainer.
5314 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5315 was supported instead of testing for mbx. Effectively a typo.
5317 15. The spa authenticator server code was not checking that the input it
5318 received was valid base64.
5320 16. The debug output line for the "set" modifier in ACLs was not showing the
5321 name of the variable that was being set.
5323 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5325 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5326 lookup on its own. Something else was needed to trigger the lookup. For
5327 example, a match in host_lookup or the need for a host name in a host list.
5328 Now, if $sender_host_name is referenced and the host name has not yet been
5329 looked up, a lookup is performed. If the lookup fails, the variable remains
5330 empty, and $host_lookup_failed is set to "1".
5332 19. Added "eqi" as a case-independent comparison operator.
5334 20. The saslauthd authentication condition could segfault if neither service
5335 nor realm was specified.
5337 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5338 error message that was logged was misleading, and incoming SMTP
5339 connections were dropped. The message is now more accurate, and temporary
5340 errors are given to SMTP connections.
5342 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5343 (except RSET) with a 421 error, until QUIT is received. However, it was
5344 failing to send a response to QUIT.
5346 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5347 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5348 ACL is now run after the helo_try_verify_hosts code.
5350 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5351 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5352 case-independent, so other case variants are also recognized). Apparently
5353 some systems use these upper case variants.
5355 25. If more than two messages were waiting for the same host, and a transport
5356 filter was specified for the transport, Exim sent two messages over the
5357 same TCP/IP connection, and then failed with "socket operation on non-
5358 socket" when it tried to send the third.
5360 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5362 27. The extern definition of crypt16() in expand.c was not being excluded when
5363 the OS had its own crypt16() function.
5365 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5366 as a preferred synonym for return_size_limit, both as an option and as an
5369 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5371 30. Changed the default configuration ACL to relax the local part checking rule
5372 for addresses that are not in any local domains. For these addresses,
5373 slashes and pipe symbols are allowed within local parts, but the sequence
5374 /../ is explicitly forbidden.
5376 31. SPA server authentication was not clearing the challenge buffer before
5379 32. log_message in a "warn" ACL statement was writing to the reject log as
5380 well as to the main log, which contradicts the documentation and doesn't
5381 seem right (because no rejection is happening). So I have stopped it.
5383 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5384 However, I am unable to do any testing of this.
5386 34. Fixed an infelicity in the appendfile transport. When checking directories
5387 for a mailbox, to see if any needed to be created, it was accidentally
5388 using path names with one or more superfluous leading slashes; tracing
5389 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5391 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5392 contents are added to the log line that is written for every discarded
5393 recipient. (Previously a log_message setting was ignored.)
5395 36. The ${quote: operator now quotes the string if it is empty.
5397 37. The install script runs exim in order to find its version number. If for
5398 some reason other than non-existence or emptiness, which it checks, it
5399 could not run './exim', it was installing it with an empty version number,
5400 i.e. as "exim-". This error state is now caught, and the installation is
5403 38. An argument was missing from the function that creates an error message
5404 when Exim fails to connect to the socket for saslauthd authentication.
5405 This could cause Exim to crash, or give a corrupted message.
5407 39. Added isip, isip4, and isip6 to ${if conditions.
5409 40. The ACL variables $acl_xx are now saved with the message, and can be
5410 accessed later in routers, transports, and filters.
5412 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
5413 strings in the file are not string-expanded.
5415 42. If a MAIL command specified a SIZE value that was too large to fit into an
5416 int variable, the check against message_size_limit failed. Such values are
5417 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
5418 day this will have to be increased, but I don't think I want to be around
5419 when emails are that large.
5426 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
5427 iconv() is not standard in FreeBSD.
5429 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
5430 IPv6 enabled. The observed symptom was a segmentation fault on return from
5431 the function os_common_find_running_interfaces() in src/os.c.
5433 3. In the check_special_case() function in daemon.c I had used "errno" as an
5434 argument name, which causes warnings on some systems. This was basically a
5435 typo, since it was named "eno" in the comments!
5437 4. The code that waits for the clock to tick (at a resolution of some fraction
5438 of a second) so as to ensure message-id uniqueness was always waiting for
5439 at least one whole tick, when it could have waited for less. [This is
5440 almost certainly not relevant at current processor speeds, where it is
5441 unlikely to ever wait at all. But we try to future-proof.]
5443 5. The function that sleeps for a time interval that includes fractions of a
5444 second contained a race. It did not block SIGALRM between setting the
5445 timer, and suspending (a couple of lines later). If the interval was short
5446 and the sigsuspend() was delayed until after it had expired, the suspension
5447 never ended. On busy systems this could lead to processes getting stuck for
5450 6. Some uncommon configurations may cause a lookup to happen in a queue runner
5451 process, before it forks any delivery processes. The open lookup caching
5452 mechanism meant that the open file or database connection was passed into
5453 the delivery process. The problem was that delivery processes always tidy
5454 up cached lookup data. This could cause a problem for the next delivery
5455 process started by the queue runner, because the external queue runner
5456 process does not know about the closure. So the next delivery process
5457 still has data in the lookup cache. In the case of a file lookup, there was
5458 no problem because closing a file descriptor in a subprocess doesn't affect
5459 the parent. However, if the lookup was caching a connection to a database,
5460 the connection was closed, and the second delivery process was likely to
5461 see errors such as "PGSQL: query failed: server closed the connection
5462 unexpectedly". The problem has been fixed by closing all cached lookups
5463 in a queue runner before running a delivery process.
5465 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
5466 seem to have the "const" qualifier which it has on other OS. I've
5469 8. Change 4.21/2 was too strict. It is only if there are two authenticators
5470 *of the same type* (client or server) with the same public name that an
5471 error should be diagnosed.
5473 9. When Exim looked up a host name for an IP address, but failed to find the
5474 original IP address when looking up the host name (a safety check), it
5475 output the message "<ip address> does not match any IP for NULL", which was
5476 confusing, to say the least. The bug was that the host name should have
5477 appeared instead of "NULL".
5479 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
5480 user that is built into the binary, and the -C or -D options is used, root
5481 privilege is dropped before the configuration file is read. In addition,
5482 logging is switched to stderr instead of the normal log files. If the
5483 configuration then re-defines the Exim user, the unprivileged environment
5484 is probably not what is expected, so Exim logs a panic warning message (but
5487 However, if deliver_drop_privilege is set, the unprivileged state may well
5488 be exactly what is intended, so the warning has been cut out in that case,
5489 and Exim is allowed to try to write to its normal log files.
5495 1. smtp_return_error_details was not giving details for temporary sender
5496 or receiver verification errors.
5498 2. Diagnose a configuration error if two authenticators have the same public
5501 3. Exim used not to create the message log file for a message until the first
5502 delivery attempt. This could be confusing when incoming messages were held
5503 for policy or load reasons. The message log file is now created at the time
5504 the message is received, and an initial "Received" line is written to it.
5506 4. The automatically generated man page for command line options had a minor
5507 bug that caused no ill effects; however, a more serious problem was that
5508 the procedure for building the man page automatically didn't always
5509 operate. Consequently, release 4.20 contains an out-of-date version. This
5510 shouldn't happen again.
5512 5. When building Exim with embedded Perl support, the script that builds the
5513 Makefile was calling 'perl' to find its compile-time parameters, ignoring
5514 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
5516 6. The freeze_tell option was not being used for messages that were frozen on
5517 arrival, either by an ACL or by local_scan().
5519 7. Added the smtp_incomplete_transaction log selector.
5521 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
5522 was accepting AUTH without a new EHLO.
5524 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
5525 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
5528 10. Logging of TCP/IP connections (when configured) now happens in the main
5529 daemon process instead of the child process, so that the TCP/IP connection
5530 count is more accurate (but it can never be perfect).
5532 11. The use of "drop" in a nested ACL was not being handled correctly in the
5533 outer ACL. Now, if condition failure induced by the nested "drop" causes
5534 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
5535 or "require"), the connection is dropped.
5537 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
5538 that yield "discard" can now be used with an "accept" or a "discard" verb,
5539 but an error is generated for any others (because I can't see a useful way
5540 to define what should happen).
5542 13. When an ACL is read dynamically from a file (or anywhere else), the lines
5543 are now processed in the same way as lines in the Exim configuration file.
5544 In particular, continuation lines are supported.
5546 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
5548 15. Added -ti meaning -t -i.
5550 16. Check for letters, digits, hyphens, and dots in the names of dnslist
5551 domains, and warn by logging if others are found.
5553 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
5554 returned from GIFCONF when Exim is trying to find the list of interfaces on
5555 a host. The code in os.c has been modified to copy each ifreq to an aligned
5556 structure in all cases.
5558 Also, in some cases, the returned ifreq's were being copied to a 'struct
5559 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
5560 means the last couple of bytes of an IPv6 address could be chopped if the
5561 ifreq contained only a normal sockaddr (14 bytes storage).
5563 18. Named domain lists were not supported in the hosts_treat_as_local option.
5564 An entry such as +xxxx was not recognized, and was treated as a literal
5567 19. Ensure that header lines added by a DATA ACL are included in the reject log
5568 if the ACL subsequently rejects the message.
5570 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
5571 MD5 (which is deprecated).
5573 21. When testing a filter file using -bf, Exim was writing a message when it
5574 took the sender from a "From " line in the message, but it was not doing so
5575 when it took $return_path from a Return-Path: header line. It now does.
5577 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
5578 with a valid header line field name (a series of printing characters
5579 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
5581 23. Changed "disc" in the source to "disk" to conform to the documentation and
5582 the book and for uniformity.
5584 24. Ignore Sendmail's -Ooption=value command line item.
5586 25. When execve() failed while trying to run a command in a pipe transport,
5587 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
5588 could be confused with a return value of 69 from the command itself. This
5589 has been changed to 127, the value the shell returns if it is asked to run
5590 a non-existent command. The wording for the related log line suggests a
5591 non-existent command as the problem.
5593 26. If received_header_text expands to an empty string, do not add a Received:
5594 header line to the message. (Well, it adds a token one on the spool, but
5595 marks it "old" so that it doesn't get used or transmitted.)
5597 27. Installed eximstats 1.28 (addition of -nt option).
5599 28. There was no check for failure on the call to getsockname() in the daemon
5600 code. This can fail if there is a shortage of resources on the system, with
5601 ENOMEM, for example. A temporary error is now given on failure.
5603 29. Contrary to the C standard, it seems that in some environments, the
5604 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
5605 program. Exim now does this explicitly; it affects the formatting of
5606 timestamps using strftime().
5608 30. If exiqsumm was given junk data, it threw up some uninitialized variable
5609 complaints. I've now initialized all the variables, to avoid this.
5611 32. Header lines added by a system filter were not being "seen" during
5612 transport-time rewrites.
5614 33. The info_callback() function passed to OpenSSL is set up with type void
5615 (*)(SSL *, int, int), as described somewhere. However, when calling the
5616 function (actually a macro) that sets it up, the type void(*)() is
5617 expected. I've put in a cast to prevent warnings from picky compilers.
5619 34. If a DNS black list lookup found a CNAME record, but there were no A
5620 records associated with the domain it pointed at, Exim crashed.
5622 35. If a DNS black list lookup returned more than one A record, Exim ignored
5623 all but the first. It now scans all returned addresses if a particular IP
5624 value is being sought. In this situation, the contents of the
5625 $dnslist_value variable are a list of all the addresses, separated by a
5628 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
5629 to accept a host name and all its aliases if the forward lookup for any of
5630 them yielded the IP address of the incoming connection. Now it accepts only
5631 those names whose forward lookup yields the correct IP address. Any other
5632 names are discarded. This closes a loophole whereby a rogue DNS
5633 administrator could create reverse DNS records to break through a
5634 wildcarded host restriction in an ACL.
5636 37. If a user filter or a system filter that ran in a subprocess used any of
5637 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
5638 the wrong values were passed to the pipe command ($thisaddress had the
5639 value of $0, $0 had the value of $1, etc). This bug was introduced by
5640 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
5642 38. Improved the line breaking for long SMTP error messages from ACLs.
5643 Previously, if there was no break point between 40 and 75 characters, Exim
5644 left the rest of the message alone. Two changes have been made: (a) I've
5645 reduced the minimum length to 35 characters; (b) if it can't find a break
5646 point between 35 and 75 characters, it looks ahead and uses the first one
5647 that it finds. This may give the occasional overlong line, but at least the
5648 remaining text gets split now.
5650 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
5651 file descriptors might be low, and that setting 1000 would always raise it.
5652 It turns out that in some environments, the limit is already over 1000 and
5653 that lowering it causes trouble. So now Exim takes care not to decrease it.
5655 40. When delivering a message, the value of $return_path is set to $sender_
5656 address at the start of routing (routers may change the value). By an
5657 oversight, this default was not being set up when an address was tested by
5658 -bt or -bv, which affected the outcome if any router or filter referred to
5661 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
5662 log only when "message" or "log_message" are set. However, if one of the
5663 conditions was an address verification, or a call to a nested ACL, the
5664 messages generated by the underlying test were being passed through. This
5665 no longer happens. The underlying message is available in $acl_verify_
5666 message for both "message" and "log_message" expansions, so it can be
5667 passed through if needed.
5669 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
5670 new expansion $bh_ to give the encoded byte string without charset
5671 translation. Translation happens only if iconv() is available; HAVE_ICONV
5672 indicates this at build time. HEADERS_CHARSET gives the charset to
5673 translate to; headers_charset can change it in the configuration, and
5674 "headers charset" can change it in an individual filter file.
5676 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
5677 that creates RFC 2047 encoded "words" labels them as that charset instead
5678 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
5679 expansion operator; (ii) when Exim creates a From: line for a local
5680 message; (iii) when a header line is rewritten to include a "phrase" part.
5682 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
5683 buggy, causing it to skip the first lines of messages whose message ID
5684 ended in 'D'. This would not have bitten before Exim release 4.14, because
5685 message IDs were unlikely to end in 'D' before then. The effect was to have
5686 incorrect size information for certain domains.
5688 45. #include "config.h" was missing at the start of the crypt16.c module. This
5689 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
5692 46. If there was a timeout during a "random" callout check, Exim treated it as
5693 a failure of the random address, and carried on sending RSET and the real
5694 address. If the delay was just some slowness somewhere, the response to the
5695 original RCPT would be taken as a response to RSET and so on, causing
5696 mayhem of various kinds.
5698 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
5699 when I implemented it. It didn't allow for the fact that some option values
5700 may legitimatetly be negative (e.g. size_addition), and it didn't even do
5701 the right test for positive values.
5703 48. Domain names in DNS records are case-independent. Exim always looks them up
5704 in lower case. Some resolvers return domain names in exactly the case they
5705 appear in the zone file, that is, they may contain uppercase letters. Not
5706 all resolvers do this - some return always lower case. Exim was treating a
5707 change of case by a resolver as a change of domain, similar to a widening
5708 of a domain abbreviation. This triggered its re-routing code and so it was
5709 trying to route what was effectively the same domain again. This normally
5710 caused routing to fail (because the router wouldn't handle the domain
5711 twice). Now Exim checks for this case specially, and just changes the
5712 casing of the domain that it ultimately uses when it transmits the message
5715 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
5718 50. If a filter generated a file delivery with a non-absolute name (possible if
5719 no home directory exists for the router), the forbid_file option was not
5722 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
5723 the existing equality matching.
5725 52. Exim was using ints instead of ino_t variables in some places where it was
5726 dealing with inode numbers.
5728 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
5729 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
5730 called TMPDIR, and if it finds it is different, it changes its value.
5732 54. The smtp_printf() function is now made available to local_scan() so
5733 additional output lines can be written before returning. There is also an
5734 smtp_fflush() function to enable the detection of a dropped connection.
5735 The variables smtp_input and smtp_batched_input are exported to
5738 55. Changed the default runtime configuration: the message "Unknown user"
5739 has been removed from the ACL, and instead placed on the localuser router,
5740 using the cannot_route_message feature. This means that any verification
5741 failures that generate their own messages won't get overridden. Similarly,
5742 the "Unrouteable address" message that was in the ACL for unverifiable
5743 relay addresses has also been removed.
5745 56. Added hosts_avoid_esmtp to the smtp transport.
5747 57. The exicyclog script was not checking for the esoteric option
5748 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
5749 will work only if exicyclog is run under the appropriate euid.
5751 58. Following a discussion on the list, the rules by which Exim recognises line
5752 endings on incoming messages have been changed. The -dropcr and drop_cr
5753 options are now no-ops, retained only for backwards compatibility. The
5754 following line terminators are recognized: LF CRLF CR. However, special
5755 processing applies to CR:
5757 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
5758 nor a local message in the state where . is a terminator.
5760 (ii) If a bare CR is encountered in a header line, an extra space is added
5761 after the line terminator so as not to end the header. The reasoning
5762 behind this is that bare CRs in header lines are most likely either
5763 to be mistakes, or people trying to play silly games.
5765 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
5766 was being incorrectly given as 18 bytes larger than it should have been.
5767 This is a VOB (very old bug).
5769 60. This may never have affected anything current, but just in case it has:
5770 When the local host is found other than at the start of a list of hosts,
5771 the local host, those with the same MX, and any that follow, are discarded.
5772 When the list in question was part of a longer list of hosts, the following
5773 hosts (not currently being processed) were also being discarded. This no
5774 longer happens. I'm not sure if this situation could ever has previously
5777 61. Added the "/MX" feature to lists of hosts in the manualroute and query
5780 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
5781 header. This is something that is recommended in a new Internet Draft, and
5782 is something that is documented as being done by Sendmail. There are two
5783 possible values. For messages generated by the autoreply transport, Exim
5786 Auto-Submitted: auto-replied
5788 whereas for all other generated messages (e.g. bounces) it adds
5790 Auto-Submitted: auto-generated
5792 63. The "personal" condition in filters now includes a test for the
5793 Auto-Submitted: header. If it contains the string "auto-" the message it
5794 not considered personal.
5796 64. Added rcpt_include_affixes as a generic transport option.
5798 65. Added queue_only_override (default true).
5800 66. Added the syslog_duplication option.
5802 67. If what should have been the first header line of a message consisted of
5803 a space followed by a colon, Exim was mis-interpreting it as a header line.
5804 It isn't of course - it is syntactically invalid and should therefore be
5805 treated as the start of the message body. The misbehaviour could have
5806 caused a number of strange effects, including loss of data in subsequent
5807 header lines, and spool format errors.
5809 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
5810 client host had authenticated. This control can now be exercised by an ACL
5811 for more flexibility.
5813 69. By default, callouts do not happen when testing with -bh. There is now a
5814 variant, -bhc, which does actually run the callout code, including
5815 consulting and updating the callout cache.
5817 70. Added support for saslauthd authentication, courtesy of Alexander
5820 71. If statvfs() failed on the spool or log directories while checking their
5821 size for availability, Exim confusingly gave the error "space shortage".
5822 Furthermore, in debugging mode it crashed with a floating point exception.
5823 These checks are done if check_{spool,log}_{space,inodes} are set, and when
5824 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
5825 serious problem, Exim now writes to the main and panic logs when this
5826 happens, with details of the failure. It then refuses to accept the
5827 incoming message, giving the message "spool directory problem" or "log
5828 directory problem" with a 421 code for SMTP messages.
5830 72. When Exim is about to re-exec itself, it ensures that the file descriptors
5831 0, 1, and 2 exist, because some OS complain for execs without them (see
5832 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
5833 descriptors. However, the code omitted to check that the open succeeded,
5834 causing mysterious errors if for some reason the permissions on /dev/null
5835 got screwed. Now Exim writes a message to the main and panic logs, and
5836 bombs out if it can't open /dev/null.
5838 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
5839 interact so that it is all more flexible. It is supposed to remain
5840 backwards compatible. Also added extra_local_interfaces.
5842 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
5843 to bomb out with an assertion failure - to the client this appears as a
5844 connection drop. This problem occurs in the part of the code that was taken
5845 from the Samba project. Fortunately, the assertion is in a very simple
5846 function, so I have fixed this by reproducing the function inline in the
5847 one place where it is called, and arranging for authentication to fail
5848 instead of killing the process with assert().
5850 75. The SPA client code was not working when the server requested OEM rather
5851 than Unicode encoding.
5853 76. Added code to make require_files with a specific uid setting more usable in
5854 the case where statting the file as root fails - usually a non-root-mounted
5855 NFS file system. When this happens and the failure is EACCES, Exim now
5856 forks a subprocess and does the per-uid checking as the relevant uid.
5858 77. Added process_log_path.
5860 78. If log_file_path was not explicitly set, a setting of check_log_space or
5861 check_log_inodes was ignored.
5863 79. If a space check for the spool or log partitions fails, the incident is now
5864 logged. Of course, in the latter case the data may get lost...
5866 80. Added the %p formatting code to string_format() so that it can be used to
5867 print addresses in debug_print(). Adjusted all the address printing in the
5868 debugging in store.c to use %p rather than %d.
5870 81. There was a concern that a line of code in smtp_in.c could overflow a
5871 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
5872 initially expressed, the concern was not well-founded, because trailing
5873 spaces are removed early. However, if the trailing spaces were followed by
5874 a NULL, they did not get removed, so the overflow was possible. Two fixes
5877 (a) I re-wrote the offending code in a cleaner fashion.
5878 (b) If an incoming SMTP command contains a NULL character, it is rejected
5881 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
5882 runs initgroups(), so that if the Exim user is in any additional groups,
5883 they will be used during message reception.
5889 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / heiko / exim.git / blob