1 Change log file for Exim from version 4.21
2 -------------------------------------------
7 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
9 PP/02 Make -n do something, by making it not do something.
10 When combined with -bP, the name of an option is not output.
12 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
15 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
16 $sender_host_name and config options to manage this, and basic check
19 PP/05 DSCP support for outbound connections and control modifier for inbound.
21 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
22 (Only plugin which currently uses this is kerberos4, which nobody should
23 be using, but we should make it available and other future plugins might
24 conceivably use it, even though it would break NAT; stuff *should* be
25 using channel bindings instead).
27 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
28 name; added for Sendmail compatibility; requires admin caller.
29 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
30 just ignore it); requires trusted caller.
31 Also parse but ignore: -Ac -Am -X<logfile>
34 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
36 TL/02 Add +smtp_confirmation as a default logging option.
38 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
39 Patch by Magnus Holmgren from 2007-02-20.
41 TL/04 Bugzilla 1281 - Spec typo.
42 Bugzilla 1283 - Spec typo.
43 Bugzilla 1290 - Spec grammar fixes.
45 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
47 TL/06 Add Experimental DMARC support using libopendmarc libraries.
49 TL/07 Fix an out of order global option causing a segfault. Reported to dev
50 mailing list by by Dmitry Isaikin.
52 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
54 JH/02 Support "G" suffix to numbers in ${if comparisons.
56 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
58 NM/01 Bugzilla 1197 - Spec typo
59 Bugzilla 1196 - Spec examples corrections
61 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
63 PP/09 Add gnutls_enable_pkcs11 option.
65 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
66 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
68 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
69 "acl {{name}{arg}...}", and optional args on acl condition
72 JH/05 Permit multiple router/transport headers_add/remove lines.
74 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
76 JH/07 Avoid using a waiting database for a single-message-only transport.
77 Performance patch from Paul Fisher. Bugzilla 1262.
79 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
82 JH/09 Add $headers_added variable, with content from use of ACL modifier
83 add_header (but not yet added to the message). Bugzilla 199.
85 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
86 Pulled from Bugzilla 817 by Wolfgang Breyha.
88 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
90 (nb: this is the same fix as in Exim 4.80.1)
92 JH/11 Add A= logging on delivery lines, and a client_set_id option on
95 JH/12 Add optional authenticated_sender logging to A= and a log_selector
98 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
100 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
101 advertise SMTP AUTH mechanism to us, instead of a generic
102 protocol violation error. Also, make Exim more robust to bad
103 data from the Dovecot auth socket.
105 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
107 When a queue runner is handling a message, Exim first routes the
108 recipient addresses, during which it prunes them based on the retry
109 hints database. After that it attempts to deliver the message to
110 any remaining recipients. It then updates the hints database using
113 So if a recipient address works intermittently, it can get repeatedly
114 deferred at routing time. The retry hints record remains fresh so the
115 address never reaches the final cutoff time.
117 This is a fairly common occurrence when a user is bumping up against
118 their storage quota. Exim had some logic in its local delivery code
119 to deal with this. However it did not apply to per-recipient defers
120 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
122 This change adds a proper retry rule check during routing so that the
123 final cutoff time is checked against the message's age. We only do
124 this check if there is an address retry record and there is not a
125 domain retry record; this implies that previous attempts to handle
126 the address had the retry_use_local_parts option turned on. We use
127 this as an approximation for the destination being like a local
128 delivery, as in LMTP.
130 I suspect this new check makes the old local delivery cutoff check
131 redundant, but I have not verified this so I left the code in place.
133 TF/02 Correct gecos expansion when From: is a prefix of the username.
135 Test 0254 submits a message to Exim with the header
139 When I ran the test suite under the user fanf2, Exim expanded
140 the header to contain my full name, whereas it should have added
141 a Resent-Sender: header. It erroneously treats any prefix of the
142 username as equal to the username.
144 This change corrects that bug.
146 GF/01 DCC debug and logging tidyup
147 Error conditions log to paniclog rather than rejectlog.
148 Debug lines prefixed by "DCC: " to remove any ambiguity.
150 TF/03 Avoid unnecessary rebuilds of lookup-related code.
152 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
153 Bug spotted by Jeremy Harris; was flawed since initial commit.
154 Would have resulted in OCSP responses post-SNI triggering an Exim
155 NULL dereference and crash.
157 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
159 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
160 Bug detection, analysis and fix by Samuel Thibault.
161 Bugzilla 1331, Debian bug #698092.
163 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
165 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
166 Server implementation by Todd Lyons, client by JH.
167 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
168 config variable "prdr_enable" controls whether the server
169 advertises the facility. If the client requests PRDR a new
170 acl_data_smtp_prdr ACL is called once for each recipient, after
171 the body content is received and before the acl_smtp_data ACL.
172 The client is controlled by bolth of: a hosts_try_prdr option
173 on the smtp transport, and the server advertisement.
174 Default client logging of deliveries and rejections involving
175 PRDR are flagged with the string "PRDR".
177 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
178 fclose(). Diagnosis by Todd Lyons.
180 PP/17 Update configure.default to handle IPv6 localhost better.
181 Patch by Alain Williams (plus minor tweaks).
184 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
185 This is now consistent with GnuTLS, and is now documented: the
186 previous undocumented portable approach to treating the option as
187 unset was to force an expansion failure. That still works, and
188 an empty string is now equivalent.
190 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
191 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
192 not performing validation itself.
194 PP/20 Added force_command boolean option to pipe transport.
195 Patch from Nick Koston, of cPanel Inc.
197 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
200 TF/04 Added udpsend ACL modifer and hexquote expansion operator
202 PP/21 Fix eximon continuous updating with timestamped log-files.
203 Broken in a format-string cleanup in 4.80, missed when I repaired the
204 other false fix of the same issue.
205 Report and fix from Heiko Schlichting.
208 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
209 Report from Prashanth Katuri.
211 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
212 It's SecureTransport, so affects any MacOS clients which use the
213 system-integrated TLS libraries, including email clients.
215 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
216 using a MIME ACL for non-SMTP local injection.
217 Report and assistance in diagnosis by Warren Baker.
219 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
221 JH/16 Fix comparisons for 64b. Bugzilla 1385.
223 TL/09 Add expansion variable $authenticated_fail_id to keep track of
224 last id that failed so it may be referenced in subsequent ACL's.
226 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
229 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
230 ldap library initialization, allowing self-signed CA's to be
231 used. Also properly sets require_cert option later in code by
232 using NULL (global ldap config) instead of ldap handle (per
233 session). Bug diagnosis and testing by alxgomz.
235 TL/12 Enhanced documentation in the ratelimit.pl script provided in
236 the src/util/ subdirectory.
238 TL/13 Bug 1301 - Imported transport SQL logging patch from Axel Rau
239 renamed to Transport Post Delivery Action by Jeremy Harris, as
246 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
248 This, or similar/improved, will also be change PP/11 of 4.82.
254 PP/01 Handle short writes when writing local log-files.
255 In practice, only affects FreeBSD (8 onwards).
256 Bugzilla 1053, with thanks to Dmitry Isaikin.
258 NM/01 Bugzilla 949 - Documentation tweak
260 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
263 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
265 PP/02 Implemented gsasl authenticator.
267 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
269 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
270 `pkg-config foo` for cflags/libs.
272 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
273 with rest of GSASL and with heimdal_gssapi.
275 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
276 `pkg-config foo` for cflags/libs for the TLS implementation.
278 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
279 properties get this fed in as external SSF. A number of robustness
280 and debugging improvements to the cyrus_sasl authenticator.
282 PP/08 cyrus_sasl server now expands the server_realm option.
284 PP/09 Bugzilla 1214 - Log authentication information in reject log.
285 Patch by Jeremy Harris.
287 PP/10 Added dbmjz lookup type.
289 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
291 PP/12 MAIL args handles TAB as well as SP, for better interop with
292 non-compliant senders.
293 Analysis and variant patch by Todd Lyons.
295 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
296 Bug report from Lars Müller <lars@samba.org> (via SUSE),
297 Patch from Dirk Mueller <dmueller@suse.com>
299 PP/13 tls_peerdn now print-escaped for spool files.
300 Observed some $tls_peerdn in wild which contained \n, which resulted
301 in spool file corruption.
303 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
304 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
305 or write after TLS renegotiation, which otherwise led to messages
308 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
309 as a tracking header (ie: a signed header comes before the signature).
310 Patch from Wolfgang Breyha.
312 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
313 comma-sep list; embedded commas doubled.
315 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
317 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
319 Report and patch from Dmitry Banschikov.
321 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
322 Removed SSL_clear() after SSL_new() which led to protocol negotiation
323 failures. We appear to now support TLS1.1+ with Exim.
325 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
326 lets Exim select keys and certificates based upon TLS SNI from client.
327 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
328 before an outbound SMTP session. New log_selector, +tls_sni.
330 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
331 NULL dereference. Report and patch from Alun Jones.
333 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
334 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
335 Not seeing resolver debug output on NetBSD, but suspect this is a
336 resolver implementation change.
338 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
339 Left warnings. Added "eximon gdb" invocation mode.
341 PP/21 Defaulting "accept_8bitmime" to true, not false.
343 PP/22 Added -bw for inetd wait mode support.
345 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
346 locate the relevant includes and libraries. Made this the default.
348 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
349 Bugzilla 1246, report and most of solution from Tomasz Kusy.
351 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
352 This may cause build issues on older platforms.
354 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
355 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
356 gnutls_require_mac & gnutls_require_protocols (no longer supported).
357 Added SNI support via GnuTLS too.
358 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
360 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
362 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
363 Applied second patch from Janne, implementing suggestion to default
364 multiple-strings-in-record handling to match SPF spec.
366 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
368 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
369 read-only, out of scope).
370 Patch from Wolfgang Breyha, report from Stuart Northfield.
372 PP/29 Fix three issues highlighted by clang analyser static analysis.
373 Only crash-plausible issue would require the Cambridge-specific
374 iplookup router and a misconfiguration.
375 Report from Marcin Mirosław.
377 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
379 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
380 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
381 As part of this, removing so much warning spew let me fix some minor
382 real issues in debug logging.
384 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
385 assignment on my part. Fixed.
387 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
388 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
389 Janne Snabb (who went above and beyond: thank you).
391 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
392 string otherwise requires a connection and a bunch more work and it's
393 relatively easy to get wrong. Should also expose TLS library linkage
396 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
397 64-bit ${eval} (JH/03).
399 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
400 GNU libc to support some of the 64-bit stuff, should not lead to
401 conflicts. Defined before os.h is pulled in, so if a given platform
402 needs to override this, it can.
404 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
405 protection layer was required, which is not implemented.
406 Bugzilla 1254, patch from Wolfgang Breyha.
408 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
409 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
410 tls_dhparam take prime identifiers. Also unbreak combination of
411 OpenSSL+DH_params+TLSSNI.
413 PP/39 Disable SSLv2 by default in OpenSSL support.
419 PP/01 Solaris build fix for Oracle's LDAP libraries.
420 Bugzilla 1109, patch from Stephen Usher.
422 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
424 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
427 TF/02 Fix a couple more cases where we did not log the error message
428 when unlink() failed. See also change 4.74-TF/03.
430 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
431 lock up or crash if it happened to be inside a call to libc when it
432 got a SIGUSR1 from exiwhat.
434 The SIGUSR1 handler appends the current process status to the process
435 log which is later printed by exiwhat. It used to use the general
436 purpose logging code to do this, but several functions it calls are
437 not safe for signals.
439 The new output code in the SIGUSR1 handler is specific to the process
440 log, and simple enough that it's easy to inspect for signal safety.
441 Removing some special cases also simplifies the general logging code.
442 Removing the spurious timestamps from the process log simplifies
445 TF/04 Improved ratelimit ACL condition.
447 The /noupdate option has been deprecated in favour of /readonly which
448 has clearer semantics. The /leaky, /strict, and /readonly update modes
449 are mutually exclusive. The update mode is no longer included in the
450 database key; it just determines when the database is updated. (This
451 means that when you upgrde Exim will forget old rate measurements.)
453 Exim now checks that the per_* options are used with an update mode that
454 makes sense for the current ACL. For example, when Exim is processing a
455 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
456 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
457 must specify per_mail/readonly. If you omit the update mode it defaults to
458 /leaky where that makes sense (as before) or /readonly where required.
460 The /noupdate option is now undocumented but still supported for
461 backwards compatibility. It is equivalent to /readonly except that in
462 ACLs where /readonly is required you may specify /leaky/noupdate or
463 /strict/noupdate which are treated the same as /readonly.
465 A useful new feature is the /count= option. This is a generalization
466 of the per_byte option, so that you can measure the throughput of other
467 aggregate values. For example, the per_byte option is now equivalent
468 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
470 The per_rcpt option has been generalized using the /count= mechanism
471 (though it's more complicated than the per_byte equivalence). When it is
472 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
473 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
474 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
475 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
476 Note that using per_rcpt with a non-readonly update mode in more than
477 one ACL will cause the recipients to be double-counted. (The per_mail
478 and per_byte options don't have this problem.)
480 The handling of very low rates has changed slightly. If the computed rate
481 is less than the event's count (usually one) then this event is the first
482 after a long gap. In this case the rate is set to the same as this event's
483 count, so that the first message of a spam run is counted properly.
485 The major new feature is a mechanism for counting the rate of unique
486 events. The new per_addr option counts the number of different
487 recipients that someone has sent messages to in the last time period. It
488 behaves like per_rcpt if all the recipient addresses are different, but
489 duplicate recipient addresses do not increase the measured rate. Like
490 the /count= option this is a general mechanism, so the per_addr option
491 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
492 example, measure the rate that a client uses different sender addresses
493 with the options per_mail/unique=$sender_address. There are further
494 details in the main documentation.
496 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
498 TF/06 Removed a few PCRE remnants.
500 TF/07 Automatically extract Exim's version number from tags in the git
501 repository when doing development or release builds.
503 PP/02 Raise smtp_cmd_buffer_size to 16kB.
504 Bugzilla 879. Patch from Paul Fisher.
506 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
507 Heavily based on revision 40f9a89a from Simon Arlott's tree.
510 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
512 PP/05 Variable $av_failed, true if the AV scanner deferred.
513 Bugzilla 1078. Patch from John Horne.
515 PP/06 Stop make process more reliably on build failure.
516 Bugzilla 1087. Patch from Heiko Schlittermann.
518 PP/07 Make maildir_use_size_file an _expandable_ boolean.
519 Bugzilla 1089. Patch from Heiko Schlittermann.
521 PP/08 Handle ${run} returning more data than OS pipe buffer size.
522 Bugzilla 1131. Patch from Holger Weiß.
524 PP/09 Handle IPv6 addresses with SPF.
525 Bugzilla 860. Patch from Wolfgang Breyha.
527 PP/10 GnuTLS: support TLS 1.2 & 1.1.
529 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
532 PP/11 match_* no longer expand right-hand-side by default.
533 New compile-time build option, EXPAND_LISTMATCH_RHS.
534 New expansion conditions, "inlist", "inlisti".
536 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
538 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
540 PP/14 fix log_write() format string regression from TF/03.
541 Bugzilla 1152. Patch from Dmitry Isaikin.
547 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
549 PP/02 Harmonised TLS library version reporting; only show if debugging.
550 Layout now matches that introduced for other libraries in 4.74 PP/03.
552 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
554 PP/04 New "dns_use_edns0" global option.
556 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
559 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
560 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
562 TK/01 Updated PolarSSL code to 0.14.2.
563 Bugzilla 1097. Patch from Andreas Metzler.
565 PP/07 Catch divide-by-zero in ${eval:...}.
568 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
571 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
572 format-string attack -- SECURITY: remote arbitrary code execution.
574 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
575 time unintentionally subject to list matching rules, letting the header
576 cause arbitrary Exim lookups (of items which can occur in lists, *not*
577 arbitrary string expansion). This allowed for information disclosure.
579 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
580 INT_MIN/-1 -- value coerced to INT_MAX.
586 NM/01 Workround for PCRE version dependency in version reporting
589 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
590 This fixes portability to compilers other than gcc, notably
591 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
593 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
594 makefiles for portability to HP-UX and POSIX correctness.
596 PP/01 Permit LOOKUP_foo enabling on the make command-line.
597 Also via indented variable definition in the Makefile.
598 (Debugging by Oliver Heesakkers).
600 PP/02 Restore caching of spamd results with expanded spamd_address.
601 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
603 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
604 Improves build reliability. Fix from: Frank Elsner
606 NM/02 Fix wide character breakage in the rfc2047 coding
607 Fixes bug 1064. Patch from Andrey N. Oktyabrski
609 NM/03 Allow underscore in dnslist lookups
610 Fixes bug 1026. Patch from Graeme Fowler
612 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
613 Code patches from Adam Ciarcinski of NetBSD.
615 NM/04 Fixed exiqgrep to cope with mailq missing size issue
618 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
619 is logged, to avoid truncation. Patch from John Horne.
621 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
622 Patch from Jakob Hirsch.
624 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
625 SQL string expansion failure details.
626 Patch from Andrey Oktyabrski.
628 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
629 Patch from Simon Arlott.
631 PP/09 New lookups functionality failed to compile on old gcc which rejects
632 extern declarations in function scope.
633 Patch from Oliver Fleischmann
635 PP/10 Use sig_atomic_t for flags set from signal handlers.
636 Check getgroups() return and improve debugging.
637 Fixed developed for diagnosis in bug 927 (which turned out to be
640 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
641 Patch from Mark Zealey.
643 PP/12 Bugzilla 1056: Improved spamd server selection.
644 Patch from Mark Zealey.
646 PP/13 Bugzilla 1086: Deal with maildir quota file races.
647 Based on patch from Heiko Schlittermann.
649 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
650 Patch from Uwe Doering, sign-off by Michael Haardt.
652 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
653 variable declaration deep within a block. Bug and patch from
656 PP/15 lookups-Makefile IRIX compatibilty coercion.
658 PP/16 Make DISABLE_DKIM build knob functional.
660 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
661 Patch by Simon Arlott
663 TF/03 Fix valgrind.h portability to C89 compilers that do not support
664 variable argument macros. Our copy now differs from upstream.
670 TF/01 Failure to get a lock on a hints database can have serious
671 consequences so log it to the panic log.
673 TF/02 Log LMTP confirmation messages in the same way as SMTP,
674 controlled using the smtp_confirmation log selector.
676 TF/03 Include the error message when we fail to unlink a spool file.
678 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
679 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
680 for maintaining out-of-tree patches for some time.
682 PP/01 Bugzilla 139: Documentation and portability issues.
683 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
684 Handle per-OS dynamic-module compilation flags.
686 PP/02 Let /dev/null have normal permissions.
687 The 4.73 fixes were a little too stringent and complained about the
688 permissions on /dev/null. Exempt it from some checks.
689 Reported by Andreas M. Kirchwitz.
691 PP/03 Report version information for many libraries, including
692 Exim version information for dynamically loaded libraries. Created
693 version.h, now support a version extension string for distributors
694 who patch heavily. Dynamic module ABI change.
696 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
697 privilege escalation vulnerability whereby the Exim run-time user
698 can cause root to append content of the attacker's choosing to
701 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
704 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
705 If dropping privileges for untrusted macros, we disabled normal logging
706 on the basis that it would fail; for the Exim run-time user, this is not
707 the case, and it resulted in successful deliveries going unlogged.
708 Fixed. Reported by Andreas Metzler.
714 PP/01 Date: & Message-Id: revert to normally being appended to a message,
715 only prepend for the Resent-* case. Fixes regression introduced in
716 Exim 4.70 by NM/22 for Bugzilla 607.
718 PP/02 Include check_rfc2047_length in configure.default because we're seeing
719 increasing numbers of administrators be bitten by this.
721 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
723 PP/03 Bugzilla 994: added openssl_options main configuration option.
725 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
727 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
729 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
731 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
732 without a peer certificate, leading to a segfault because of an
733 assumption that peers always have certificates. Be a little more
734 paranoid. Problem reported by Martin Tscholak.
736 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
737 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
738 NB: ClamAV planning to remove STREAM in "middle of 2010".
739 CL also introduces -bmalware, various -d+acl logging additions and
740 more caution in buffer sizes.
742 PP/09 Implemented reverse_ip expansion operator.
744 PP/10 Bugzilla 937: provide a "debug" ACL control.
746 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
748 PP/12 Bugzilla 973: Implement --version.
750 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
752 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
754 PP/15 Bugzilla 816: support multiple condition rules on Routers.
756 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
757 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
758 ignore trailing whitespace.
760 JJ/02 prevent non-panic DKIM error from being sent to paniclog
762 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
765 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
766 Notification from Dr Andrew Aitchison.
768 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
769 ExtendedDetectionInfo response format.
770 Notification from John Horne.
772 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
775 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
776 XSL and documented dependency on system catalogs, with examples of how
779 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
782 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
783 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
784 configuration file which is writeable by the Exim user or group.
786 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
787 of configuration files to cover files specified with the -C option if
788 they are going to be used with root privileges, not just the default
791 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
792 option (effectively making it always true).
794 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
795 files to be used while preserving root privileges.
797 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
798 that rogue child processes cannot use them.
800 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
801 run-time user, instead of root.
803 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
804 Exim run-time user without dropping privileges.
806 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
807 result string, instead of calling string_vformat() twice with the same
810 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
811 for other users. Others should always drop root privileges if they use
812 -C on the command line, even for a whitelisted configure file.
814 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
816 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
822 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
823 $data_path, and $header_path variables; fixed documentation bugs and
826 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
827 exipick to access non-standard spools, including the "frozen" queue
830 NM/01 Bugzilla 965: Support mysql stored procedures.
831 Patch from Alain Williams
833 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
835 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
836 Patch from Andreas Metzler
838 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
839 Patch from Kirill Miazine
841 NM/05 Bugzilla 671: Added umask to procmail example.
843 JJ/03 installed exipick 20100323.0, fixing doc bug
845 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
846 directory. Notification and patch from Dan Rosenberg.
848 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
850 TK/02 Improve log output when DKIM signing operation fails.
852 MH/01 Treat the transport option dkim_domain as a colon separated
853 list, not as a single string, and sign the message with each element,
854 omitting multiple occurences of the same signer.
856 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
857 Bugzilla 985, 986. Patch by Simon Arlott
859 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
860 Patch by Simon Arlott
862 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
863 MBX locking. Notification from Dan Rosenberg.
869 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
871 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
873 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
875 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
877 NM/04 Bugzilla 915: Flag broken perl installation during build.
883 TK/01 Added patch by Johannes Berg that expands the main option
884 "spamd_address" if it starts with a dollar sign.
886 TK/02 Write list of recipients to X-Envelope-Sender header when building
887 the mbox-format spool file for content scanning (suggested by Jakob
890 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
891 (http://www.dcc-servers.net/) support via dccifd. Activated by
892 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
894 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
895 by Mark Daniel Reidel <mr@df.eu>.
897 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
898 When building exim an external PCRE library is now needed -
899 PCRE is a system library on the majority of modern systems.
900 See entry on PCRE_LIBS in EDITME file.
902 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
903 conversation. Added nologin parameter to request.
904 Patch contributed by Kirill Miazine.
906 TF/01 Do not log submission mode rewrites if they do not change the address.
908 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
910 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
911 log files in place. Contributed by Roberto Lima.
913 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
915 TF/03 Bugzilla 615: When checking the local_parts router precondition
916 after a local_part_suffix or local_part_prefix option, Exim now
917 does not use the address's named list lookup cache, since this
918 contains cached lookups for the whole local part.
920 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
921 Robert Millan. Documentation is in experimental-spec.txt.
923 TF/04 Bugzilla 668: Fix parallel build (make -j).
925 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
927 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
928 Patch provided by Jan Srzednicki.
930 TF/05 Leading white space used to be stripped from $spam_report which
931 wrecked the formatting. Now it is preserved.
933 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
934 that they are available at delivery time.
936 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
938 TF/08 TLS error reporting now respects the incoming_interface and
939 incoming_port log selectors.
941 TF/09 Produce a more useful error message if an SMTP transport's hosts
942 setting expands to an empty string.
944 NM/06 Bugzilla 744: EXPN did not work under TLS.
945 Patch provided by Phil Pennock.
947 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
948 Patch provided by Richard Godbee.
950 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
951 acl_smtp_notquit, added index entry.
953 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
954 Patch provided by Eugene Bujak.
956 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
957 accept(). Patch provided by Maxim Dounin.
959 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
960 Patch provided by Phil Pennock.
962 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
964 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
965 Patch provided by Brad "anomie" Jorsch.
967 NM/14 Bugzilla 622: Added timeout setting to transport filter.
968 Patch provided by Dean Brooks.
970 TK/05 Add native DKIM support (does not depend on external libraries).
972 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
973 Patch provided by Graeme Fowler.
975 NM/16 Bugzilla 851: Documentation example syntax fix.
977 NM/17 Changed NOTICE file to remove references to embedded PCRE.
979 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
982 NM/19 Bugzilla 745: TLS version reporting.
983 Patch provided by Phil Pennock.
985 NM/20 Bugzilla 167: bool: condition support.
986 Patch provided by Phil Pennock.
988 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
989 clients. Patch provided by Phil Pennock.
991 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
992 Patch provided by Brad "anomie" Jorsch.
994 NM/23 Bugzilla 687: Fix misparses in eximstats.
995 Patch provided by Heiko Schlittermann.
997 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
998 Patch provided by Heiko Schlittermann.
1000 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1001 plus update to original patch.
1003 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1005 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1006 Patch provided by David Brownlee.
1008 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1010 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1012 NM/30 Bugzilla 888: TLS documentation bugfixes.
1014 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1016 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1017 Unlike the original bugzilla I have changed all shell scripts in src tree.
1019 NM/33 Bugzilla 898: Transport filter timeout fix.
1020 Patch by Todd Rinaldo.
1022 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1023 Patch by Serge Demonchaux.
1025 NM/35 Bugzilla 39: Base64 decode bug fixes.
1026 Patch by Jakob Hirsch.
1028 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1030 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1032 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1034 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1040 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1041 ALT-N's libdkim that I have put here:
1042 http://duncanthrax.net/exim-experimental/
1044 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1045 were called 'true' and it seems that C99 defines that as a reserved
1046 keyword to be used with 'bool' variable types. That means you could
1047 not include C99-style headers which use bools without triggering
1048 build errors in sieve.c.
1050 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1051 as mailq or other aliases. Changed the --help handling significantly
1052 to do whats expected. exim_usage() emits usage/help information.
1054 SC/01 Added the -bylocaldomain option to eximstats.
1056 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1058 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1060 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1066 PH/01 Another patch from the Sieve maintainer.
1068 PH/02 When an IPv6 address is converted to a string for single-key lookup
1069 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1070 dots are used instead of colons so that keys in lsearch files need not
1071 contain colons. This was done some time before quoting was made available
1072 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1073 (notated using the quote facility) so as to distinguish them from IPv4
1074 keys. This meant that lookups for IP addresses in host lists did not work
1075 for iplsearch lookups.
1077 This has been fixed by arranging for IPv6 addresses to be expressed with
1078 colons if the lookup type is iplsearch. This is not incompatible, because
1079 previously such lookups could never work.
1081 The situation is now rather anomolous, since one *can* have colons in
1082 ordinary lsearch keys. However, making the change in all cases is
1083 incompatible and would probably break a number of configurations.
1085 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1088 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1089 conversion specification without a maximum field width, thereby enabling
1090 a rogue spamd server to cause a buffer overflow. While nobody in their
1091 right mind would setup Exim to query an untrusted spamd server, an
1092 attacker that gains access to a server running spamd could potentially
1093 exploit this vulnerability to run arbitrary code as the Exim user.
1095 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1096 $primary_hostname instead of what libspf2 thinks the hosts name is.
1098 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1099 a directory entry by the name of the lookup key. Previously, if a
1100 symlink pointed to a non-existing file or a file in a directory that
1101 Exim lacked permissions to read, a lookup for a key matching that
1102 symlink would fail. Now it is enough that a matching directory entry
1103 exists, symlink or not. (Bugzilla 503.)
1105 PH/03 The body_linecount and body_zerocount variables are now exported in the
1108 PH/04 Added the $dnslist_matched variable.
1110 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1111 This means they are set thereafter only if the connection becomes
1114 PH/06 Added the client_condition to authenticators so that some can be skipped
1115 by clients under certain conditions.
1117 PH/07 The error message for a badly-placed control=no_multiline_responses left
1118 "_responses" off the end of the name.
1120 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1122 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1123 (without spaces) instead of just copying the configuration text.
1125 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1127 PH/11 Added $max_received_linelength.
1129 PH/12 Added +ignore_defer and +include_defer to host lists.
1131 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1132 way in which PCRE > 7.0 is built.
1134 PH/14 Implemented queue_only_load_latch.
1136 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1137 MAIL command. The effect was to mangle the value on 64-bit systems.
1139 PH/16 Another patch from the Sieve maintainer.
1141 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1143 PH/18 If a system quota error occurred while trying to create the file for
1144 a maildir delivery, the message "Mailbox is full" was not appended to the
1145 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1146 only to a quota excession during the actual writing of the file.
1148 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1149 characters?) which causes problems in log lines. The DN values are now
1150 passed through string_printing() before being added to log lines.
1152 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1153 and InterBase are left for another time.)
1155 PH/21 Added message_body_newlines option.
1157 PH/22 Guard against possible overflow in moan_check_errorcopy().
1159 PH/23 POSIX allows open() to be a macro; guard against that.
1161 PH/24 If the recipient of an error message contained an @ in the local part
1162 (suitably quoted, of course), incorrect values were put in $domain and
1163 $local_part during the evaluation of errors_copy.
1169 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1170 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1173 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1174 issue a MAIL command.
1176 PH/02 In an ACL statement such as
1178 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1180 if a client was not listed at all, or was listed with a value other than
1181 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1182 the condition was not true (as it should be), so access was not denied.
1183 The bug was that the ! inversion was incorrectly passed on to the second
1184 item. This has been fixed.
1186 PH/03 Added additional dnslists conditions == and =& which are different from
1187 = and & when the dns lookup returns more than one IP address.
1189 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1190 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1192 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1193 FSYNC, which compiles an option called disable_fsync that allows for
1194 bypassing fsync(). The documentation is heavily laced with warnings.
1196 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1198 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1199 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1200 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1201 including adding "make clean"; (3) Added -fPIC when compiling the test
1202 dynamically loaded module, to get rid of a warning.
1204 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1205 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1206 = 0s. The bug is otherwise harmless.
1208 PH/07 There was a bug in the dovecot authenticator such that the value of
1209 $auth1 could be overwritten, and so not correctly preserved, after a
1210 successful authentication. This usually meant that the value preserved by
1211 the server_setid option was incorrect.
1213 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1215 PH/09 Installed PCRE release 7.0.
1217 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1218 run for batched SMTP input. It is now run at the start of every message
1219 in the batch. While fixing this I discovered that the process information
1220 (output by running exiwhat) was not always getting set for -bs and -bS
1221 input. This is fixed, and it now also says "batched" for BSMTP.
1223 PH/11 Added control=no_pipelining.
1225 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1226 patch, slightly modified), and move the expansion of helo_data till after
1227 the connection is made in the smtp transport (so it can use these
1230 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1232 PH/14 Added log_selector = +pid.
1234 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1236 PH/16 Add ${if forany and ${if forall.
1238 PH/17 Added dsn_from option to vary the From: line in DSNs.
1240 PH/18 Flush SMTP output before performing a callout, unless control =
1241 no_callout_flush is set.
1243 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1244 was true (the default) a successful delivery failed to delete the retry
1245 item, thus causing premature timeout of the address. The bug is now
1248 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1250 PH/21 Long custom messages for fakedefer and fakereject are now split up
1251 into multiline reponses in the same way that messages for "deny" and
1252 other ACL rejections are.
1254 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1255 with slight modification.
1257 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1258 draft, changing the syntax and factoring some duplicate code".
1260 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1261 for deliveries of the second and subsequent messages over the same SMTP
1264 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1265 ${reduce, with only minor "tidies".
1267 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1269 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1270 expansion side effects.
1272 PH/27 When a message times out after an over-quota error from an Exim-imposed
1273 quota, the bounce message says "mailbox is full". This message was not
1274 being given when it was a system quota that was exceeded. It now should
1277 MH/03 Made $recipients available in local_scan(). local_scan() already has
1278 better access to the recipient list through recipients_list[], but
1279 $recipients can be useful in postmaster-provided expansion strings.
1281 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1282 in the case of a MAIL command with additional options following the
1283 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1284 were accidentally chopped off.
1286 PH/29 SMTP synchronization checks are implemented when a command is read -
1287 there is a check that no more input is waiting when there shouldn't be
1288 any. However, for some commands, a delay in an ACL can mean that it is
1289 some time before the response is written. In this time, more input might
1290 arrive, invalidly. So now there are extra checks after an ACL has run for
1291 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1292 pipelining has not been advertised.
1294 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1296 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1297 returned in the SMTP response when smtp_return_error_details was false.
1298 This has been fixed.
1300 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1301 instead of the C I/O that was originally supplied, because problems were
1302 reported on Solaris.
1304 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1305 Exim which did not show up earlier: it was assuming that a call to
1306 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1307 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1308 was a macro that became an assignment, so it seemed to work. This has
1309 changed to a proper function call with a void return, hence the compile
1310 error. Exim's code has been fixed.
1312 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1315 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1317 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1319 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1320 because it turns out to be too big or there is a timeout) there is no
1321 "Completed" line in the log. When some messages of this type were
1322 selected by exigrep, they were listed as "not completed". Others were
1323 picked up by some special patterns. I have improved the selection
1324 criteria to be more general.
1326 PH/38 The host_find_failed option in the manualroute router can now be set
1327 to "ignore", to completely ignore a host whose IP address cannot be
1328 found. If all hosts are ignored, the behaviour is controlled by the new
1329 host_all_ignored option.
1331 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1332 homing or because of multiple MX records with /mx) generated more than
1333 one IP address, and the following item turned out to be the local host,
1334 all the secondary addresses of the first item were incorrectly removed
1335 from the list, along with the local host and any following hosts (which
1336 is what is supposed to happen).
1338 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1339 whoever called Exim into the -H file. In the case of the daemon it was
1340 behaving confusingly. When first started, it used values for whoever
1341 started the daemon, but after a SIGHUP it used the Exim user (because it
1342 calls itself on a restart). I have changed the code so that it now always
1345 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1346 message are rejected with the same error (e.g. no authentication or bad
1347 sender address), and a DATA command is nevertheless sent (as can happen
1348 with PIPELINING or a stupid MUA), the error message that was given to the
1349 RCPT commands is included in the rejection of the DATA command. This is
1350 intended to be helpful for MUAs that show only the final error to their
1353 PH/42 Another patch from the Sieve maintainer.
1355 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1356 Eximstats - Fixed some broken HTML links and added missing column headers
1358 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1359 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1361 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1363 PH/43 Yet another patch from the Sieve maintainer.
1365 PH/44 I found a way to check for a TCP/IP connection going away before sending
1366 the response to the final '.' that terminates a message, but only in the
1367 case where the client has not sent further data following the '.'
1368 (unfortunately, this is allowed). However, in many cases there won't be
1369 any further data because there won't be any more messages to send. A call
1370 to select() can be used: if it shows that the input is "ready", there is
1371 either input waiting, or the socket has been closed. An attempt to read
1372 the next input character can distinguish the two cases. Previously, Exim
1373 would have sent an OK response which the client would never have see.
1374 This could lead to message repetition. This fix should cure that, at
1375 least in a lot of common cases.
1377 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1378 advertised in response to EHLO.
1384 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1385 fixed by 4.65/MH/01 (is this a record?) are fixed:
1387 (i) An empty string was always treated as zero by the numeric comparison
1388 operators. This behaviour has been restored.
1390 (ii) It is documented that the numeric comparison operators always treat
1391 their arguments as decimal numbers. This was broken in that numbers
1392 starting with 0 were being interpreted as octal.
1394 While fixing these problems I realized that there was another issue that
1395 hadn't been noticed. Values of message_size_limit (both the global option
1396 and the transport option) were treated as octal if they started with 0.
1397 The documentation was vague. These values are now always treated as
1398 decimal, and I will make that clear in the documentation.
1404 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1405 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1408 MH/01 Don't check that the operands of numeric comparison operators are
1409 integers when their expansion is in "skipping" mode (fixes bug
1410 introduced by 4.64-PH/07).
1412 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1413 child addresses, Exim now panics and dies. Previously, because the count
1414 is held in a short int, deliveries were likely to be lost. As such a
1415 large number of recipients for a single message is ridiculous
1416 (performance will be very, very poor), I have chosen to impose a limit
1417 rather than extend the field.
1423 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1424 leftover -K file (the existence of which was triggered by #402).
1425 While we were at it, introduced process PID as part of the -K
1426 filename. This should rule out race conditions when creating
1429 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1430 processing considerably. Previous code took too long for large mails,
1431 triggering a timeout which in turn triggers #401.
1433 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1434 in the DK code in transports.c. sendfile() is not really portable,
1435 hence the _LINUX specificness.
1437 TF/01 In the add_headers option to the mail command in an Exim filter,
1438 there was a bug that Exim would claim a syntax error in any
1439 header after the first one which had an odd number of characters
1442 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1443 callout verification, Exim cached a "reject" for the entire domain. This
1444 is correct for most verifications, but it is not correct for a recipient
1445 verification with use_sender or use_postmaster set, because in that case
1446 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1447 case of MAIL FROM:<> rejection from other early rejections (e.g.
1448 rejection of HELO). When verifying a recipient using a non-null MAIL
1449 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1450 Whatever the result of the callout, the value of the domain cache is
1451 left unchanged (for any other kind of callout, getting as far as trying
1452 RCPT means that the domain itself is ok).
1454 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
1457 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
1458 manifest itself as EPIPE rather than ECONNECT. When tidying away a
1459 session, the daemon ignores ECONNECT errors and logs others; it now
1460 ignores EPIPE as well.
1462 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
1463 (quoted-printable decoding).
1465 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
1466 later the small subsequent patch to fix an introduced bug.
1468 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
1470 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
1472 PH/08 An error is now given if message_size_limit is specified negative.
1474 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
1475 to be given (somewhat) arbitrary names.
1477 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
1480 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
1481 miscellaneous code fixes
1483 PH/10 Added the log_reject_target ACL modifier to specify where to log
1486 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
1487 hostname. This is wrong, because it relates to the incoming message (and
1488 probably the interface on which it is arriving) and not to the outgoing
1489 callout (which could be using a different interface). This has been
1490 changed to use the value of the helo_data option from the smtp transport
1491 instead - this is what is used when a message is actually being sent. If
1492 there is no remote transport (possible with a router that sets up host
1493 addresses), $smtp_active_hostname is used.
1495 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
1496 tweaks were necessary in order to get it to work (see also 21 below):
1497 (a) The code assumed that strncpy() returns a negative number on buffer
1498 overflow, which isn't the case. Replaced with Exim's string_format()
1500 (b) There were several signed/unsigned issues. I just did the minimum
1501 hacking in of casts. There is scope for a larger refactoring.
1502 (c) The code used strcasecmp() which is not a standard C function.
1503 Replaced with Exim's strcmpic() function.
1504 (d) The code set only $1; it now sets $auth1 as well.
1505 (e) A simple test gave the error "authentication client didn't specify
1506 service in request". It would seem that Dovecot has changed its
1507 interface. Fortunately there's a specification; I followed it and
1508 changed what the client sends and it appears to be working now.
1510 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
1513 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
1514 address A is aliased to B and C, where B exists and C does not. Without
1515 -v the output is "A verified" because verification stops after a
1516 successful redirection if more than one address is generated. However,
1517 with -v the child addresses are also verified. Exim was outputting "A
1518 failed to verify" and then showing the successful verification for C,
1519 with its parentage. It now outputs "B failed to verify", showing B's
1520 parentage before showing the successful verification of C.
1522 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
1523 look up a TXT record in a specific list after matching in a combined
1526 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
1527 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
1528 they consult the DNS. I had assumed they would set it the way they
1529 wanted; and indeed my experiments on Linux seem to show that in some
1530 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
1531 To be on the safe side, however, I have now made the interface to
1532 host_find_byname() similar to host_find_bydns(), with an argument
1533 containing the DNS resolver options. The host_find_byname() function now
1534 sets these options at its start, just as host_find_bydns() does. The smtp
1535 transport options dns_qualify_single and dns_search_parents are passed to
1536 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
1537 of host_find_byname() use the default settings of RES_DEFNAMES
1538 (qualify_single) but not RES_DNSRCH (search_parents).
1540 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
1541 spool_read_header() do less string testing, by means of a preliminary
1542 switch on the second character of optional "-foo" lines. (This is
1543 overdue, caused by the large number of possibilities that now exist.
1544 Originally there were few.) While I was there, I also converted the
1545 str(n)cmp tests so they don't re-test the leading "-" and the first
1546 character, in the hope this might squeeze out yet more improvement.
1548 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
1549 flag allowing group syntax was set by the header_syntax check but not
1550 turned off, possible causing trouble later; (2) The flag was not being
1551 set at all for the header_verify test, causing "group"-style headers to
1552 be rejected. I have now set it in this case, and also caused header_
1553 verify to ignore an empty address taken from a group. While doing this, I
1554 came across some other cases where the code for allowing group syntax
1555 while scanning a header line wasn't quite right (mostly, not resetting
1556 the flag correctly in the right place). These bugs could have caused
1557 trouble for malformed header lines. I hope it is now all correct.
1559 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
1560 with the "reply" argument non-NULL. The code, however (which originally
1561 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
1562 but it didn't always do it. This confused somebody who was copying the
1563 code for some other use. I have removed all the tests.
1565 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
1566 feature that was used to support insecure browsers during the U.S. crypto
1567 embargo. It requires special client support, and Exim is probably the
1568 only MTA that supported it -- and would never use it because real RSA is
1569 always available. This code has been removed, because it had the bad
1570 effect of slowing Exim down by computing (never used) parameters for the
1571 RSA_EXPORT functionality.
1573 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
1574 authenticator to fail if there's a tab character in the incoming data
1575 (there should never be unless someone is messing about, as it's supposed
1576 to be base64-encoded). Also added, on Timo's advice, the "secured" option
1577 if the connection is using TLS or if the remote IP is the same as the
1578 local IP, and the "valid-client-cert option" if a client certificate has
1581 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
1582 authenticators. This can be used for authorization after authentication
1583 succeeds. (In the case of plaintext, it servers for both authentication
1586 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
1587 if any retry times were supplied.
1589 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
1590 connection, where there is no client IP address to check. In this
1591 situation, the verify now always succeeds.
1593 PH/25 Applied John Jetmore's -Mset patch.
1595 PH/26 Added -bem to be like -Mset, but loading a message from a file.
1597 PH/27 In a string expansion for a processed (not raw) header when multiple
1598 headers of the same name were present, leading whitespace was being
1599 removed from all of them, but trailing whitespace was being removed only
1600 from the last one. Now trailing whitespace is removed from each header
1601 before concatenation. Completely empty headers in a concatenation (as
1602 before) are ignored.
1604 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
1605 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
1607 PH/29 [Removed. This was a change that I later backed out, and forgot to
1608 correct the ChangeLog entry (that I had efficiently created) before
1609 committing the later change.]
1611 PH/30 Exim was sometimes attempting to deliver messages that had suffered
1612 address errors (4xx response to RCPT) over the same connection as other
1613 messages routed to the same hosts. Such deliveries are always "forced",
1614 so retry times are not inspected. This resulted in far too many retries
1615 for the affected addresses. The effect occurred only when there were more
1616 hosts than the hosts_max_try setting in the smtp transport when it had
1617 the 4xx errors. Those hosts that it had tried were not added to the list
1618 of hosts for which the message was waiting, so if all were tried, there
1619 was no problem. Two fixes have been applied:
1621 (i) If there are any address or message errors in an SMTP delivery, none
1622 of the hosts (tried or untried) are now added to the list of hosts
1623 for which the message is waiting, so the message should not be a
1624 candidate for sending over the same connection that was used for a
1625 successful delivery of some other message. This seems entirely
1626 reasonable: after all the message is NOT "waiting for some host".
1627 This is so "obvious" that I'm not sure why it wasn't done
1628 previously. Hope I haven't missed anything, but it can't do any
1629 harm, as the worst effect is to miss an optimization.
1631 (ii) If, despite (i), such a delivery is accidentally attempted, the
1632 routing retry time is respected, so at least it doesn't keep
1633 hammering the server.
1635 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
1636 in ${readsocket because some servers need this prod.
1638 PH/32 Added some extra debug output when updating a wait-xxx database.
1640 PH/33 The hint "could be header name not terminated by colon", which has been
1641 given for certain expansion errors for a long time, was not being given
1642 for the ${if def:h_colon_omitted{... case.
1644 PH/34 The spec says: "With one important exception, whenever a domain list is
1645 being scanned, $domain contains the subject domain." There was at least
1646 one case where this was not true.
1648 PH/35 The error "getsockname() failed: connection reset by peer" was being
1649 written to the panic log as well as the main log, but it isn't really
1650 panic-worthy as it just means the connection died rather early on. I have
1651 removed the panic log writing for the ECONNRESET error when getsockname()
1654 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
1655 runs only) independently of the message's sender address. This meant
1656 that, if the 4xx error was in fact related to the sender, a different
1657 message to the same recipient with a different sender could confuse
1658 things. In particualar, this can happen when sending to a greylisting
1659 server, but other circumstances could also provoke similar problems.
1660 I have changed the default so that the retry time for these errors is now
1661 based a combination of the sender and recipient addresses. This change
1662 can be overridden by setting address_retry_include_sender=false in the
1665 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
1666 remote server are returned as part of bounce messages. This was not
1667 happening for LMTP over a pipe (the lmtp transport), but now it is the
1668 same for both kinds of LMTP.
1670 PH/38 Despite being documented as not happening, Exim was rewriting addresses
1671 in header lines that were in fact CNAMEs. This is no longer the case.
1673 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
1674 and queue runs started by the daemon processed all messages. This has
1675 been fixed so that -R and -S can now usefully be given with -q<time>.
1677 PH/40 Import PCRE release 6.7 (fixes some bugs).
1679 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
1681 PH/42 Give an error if -q is specified more than once.
1683 PH/43 Renamed the variables $interface_address and $interface_port as
1684 $received_ip_address and $received_port, to make it clear that these
1685 values apply to message reception, and not to the outgoing interface when
1686 a message is delivered. (The old names remain recognized, of course.)
1688 PH/44 There was no timeout on the connect() call when using a Unix domain
1689 socket in the ${readsocket expansion. There now is.
1691 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
1692 be meaningful with "accept".
1694 SC/01 Eximstats V1.43
1695 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
1697 SC/02 Eximstats V1.44
1698 Use a glob alias rather than an array ref in the generated
1699 parser. This improves both readability and performance.
1701 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
1702 Collect SpamAssassin and rejection statistics.
1703 Don't display local sender or destination tables unless
1704 there is data to show.
1705 Added average volumes into the top table text output.
1707 SC/04 Eximstats V1.46
1708 Collect data on the number of addresses (recipients)
1709 as well as the number of messages.
1711 SC/05 Eximstats V1.47
1712 Added 'Message too big' to the list of mail rejection
1713 reasons (thanks to Marco Gaiarin).
1715 SC/06 Eximstats V1.48
1716 Mainlog lines which have GMT offsets and are too short to
1717 have a flag are now skipped.
1719 SC/07 Eximstats V1.49 (Alain Williams)
1720 Added the -emptyok flag.
1722 SC/08 Eximstats V1.50
1723 Fixes for obtaining the IP address from reject messages.
1725 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
1726 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
1727 whitesspace changes from 4.64-PH/27
1729 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
1732 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
1733 are found, allow negative numbers in numeric criteria)
1735 JJ/06 exipick.20061117.2, added new $message_body_missing variable
1737 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
1738 to match changes made in 4.64-PH/43
1740 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
1742 PH/47 Put in an explicit test for a DNS lookup of an address record where the
1743 "domain" is actually an IP address, and force a failure. This locks out
1744 those revolvers/nameservers that support "A-for-A" lookups, in
1745 contravention of the specifications.
1747 PH/48 When a host name was looked up from an IP address, and the subsequent
1748 forward lookup of the name timed out, the host name was left in
1749 $sender_host_name, contrary to the specification.
1751 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
1752 restricted to single-key lookups, Exim was not diagnosing an error if
1753 * or *@ was used with a query-style lookup.
1755 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
1757 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
1758 long ago, but noone interested enough thought of it. Let's just say that
1759 the "1.1" means that there are some new functions that weren't there at
1760 some point in the past.
1762 PH/51 Error processing for expansion failure of helo_data from an smtp
1763 transport during callout processing was broken.
1765 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
1766 tested/used via the -bh/-bhc/-bs options.
1768 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
1769 bug, fixed in subsequent PCRE releases).
1771 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
1772 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
1774 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
1780 SC/01 Use a glob alias rather than an array ref in eximstats generated
1781 parser. This improves both readability and performance.
1783 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
1784 Don't display local sender or destination tables in eximstats unless
1785 there is data to show.
1786 Added average volumes into the eximstats top table text output.
1788 SC/03 Collect data on the number of addresses (recipients) as well
1789 as the number of messages in eximstats.
1791 TF/01 Correct an error in the documentation for the redirect router. Exim
1792 does (usually) call initgroups() when daemonizing.
1794 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
1795 with consistent privilege compared to when running as a daemon.
1797 TF/03 Note in the spec that $authenticated_id is not set for local
1798 submissions from trusted users.
1800 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
1801 Thanks to Dean Brooks <dean@iglou.com> for the patch.
1803 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
1804 by adding some example configuration directives to the default
1805 configuration file. A little bit of work is required to uncomment the
1806 directives and define how usernames and passwords are checked, but
1807 there is now a framework to start from.
1809 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
1810 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
1811 without this. I don't know how relevant this is to other LDAP libraries.
1813 PH/02 Add the verb name to the "unknown ACL verb" error.
1815 PH/03 Magnus Holmgren's patch for filter_prepend_home.
1817 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
1819 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
1820 directory not expanded when it should be if an expanded home directory
1821 was set for the address (which is overridden by the transport).
1823 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
1826 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
1827 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
1828 because it is too late at that time, and has no effect.
1830 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
1831 security issue with \' (bugzilla #107). I could not use the
1832 PQescapeStringConn() function, because it needs a PGconn value as one of
1835 PH/08 When testing addresses using -bt, indicate those final addresses that
1836 are duplicates that would not cause an additional delivery. At least one
1837 person was confused, thinking that -bt output corresponded to deliveries.
1838 (Suppressing duplicates isn't a good idea as you lose the information
1839 about possibly different redirections that led to the duplicates.)
1841 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
1842 systems where poll() doesn't work, in particular OS X.
1844 PH/10 Added more information to debugging output for retry time not reached.
1846 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
1847 operations in malware.c.
1849 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
1852 PH/13 If write_rejectlog was set false when logging was sent to syslog with
1853 syslog_duplication set false, log lines that would normally be written
1854 both the the main log and to the reject log were not written to syslog at
1857 PH/14 In the default configuration, change the use of "message" in ACL warn
1858 statements to "add_header".
1860 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
1861 not followed by a command (e.g. "seen endif").
1863 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
1864 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
1867 PH/17 Added extra conditions to the default value of delay_warning_condition
1871 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
1872 { match{$h_precedence:}{(?i)bulk|list|junk} } \
1873 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
1876 The Auto-Submitted: and various List- headers are standardised, whereas I
1877 don't think Precedence: ever was.
1879 PH/18 Refactored debugging code in route_finduser() to show more information,
1880 in particular, the error code if getpwnam() issues one.
1882 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
1883 This is apparently needed in addition to the PH/07 change above to avoid
1884 any possible encoding problems.
1886 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
1887 but not after initializing Perl.
1889 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
1890 output them only if debugging. By default they are written stderr,
1891 apparently, which is not desirable.
1893 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
1896 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
1899 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1901 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1902 authenticated or an ident call has been made. Suppress the default
1903 values for $authenticated_id and $authenticated_sender (but permit -oMai
1904 and -oMas) when testing with -bh.
1906 PH/24 Re-jigged the order of the tests in the default configuration so that the
1907 tests for valid domains and recipients precede the DNS black list and CSA
1908 tests, on the grounds that those ones are more expensive.
1910 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1911 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1912 This bug exists in every version of Exim that I still have, right back to
1915 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1916 However, an attempt to turn on case-sensitivity in a regex key by
1917 including (?-i) didn't work because the subject string was already
1918 lowercased, and the effects were non-intuitive. It turns out that a
1919 one-line patch can be used to allow (?-i) to work as expected.
1925 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1926 other effects) broke the use of negated acl sub-conditions.
1928 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1931 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1932 "Deny" causes Exim to reject the incoming connection with a 554 error.
1933 Unfortunately, if there is a major crisis, such as a disk failure,
1934 tcp-wrappers gives "deny", whereas what one would like would be some
1935 kind of temporary error. A kludge has been added to help with this.
1936 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1937 554 error is used if errno is still zero or contains ENOENT (which occurs
1938 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1941 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1943 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1944 errors. Otherwise a message that provokes a temporary error (when other
1945 messages do not) can cause a whole host to time out.
1947 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1948 addresses were routed directly to files or pipes from a redirect router.
1949 File deliveries just didn't batch; pipe deliveries might have suffered
1952 PH/06 A failure to get a lock for a hints database would erroneously always say
1953 "Failed to get write lock", even when it was really a read lock.
1955 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1956 of 0600. This has been changed to use the value of the lockfile_mode
1957 option (which defaults to 0600).
1959 PH/08 Applied small patch from the Sieve maintainer.
1961 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1962 folder from quota calculations, a direct delivery into this folder messed
1963 up the contents of the maildirsize file. This was because the regex was
1964 used only to exclude .Trash (or whatever) when the size of the mailbox
1965 was calculated. There was no check that a delivery was happening into an
1966 excluded directory. This bug has been fixed by ignoring all quota
1967 processing for deliveries into excluded directories.
1969 PH/10 Added the maildirfolder_create_regex option to appendfile.
1975 PH/01 The code for finding all the local interface addresses on a FreeBSD
1976 system running IPv6 was broken. This may well have applied to all BSD
1977 systems, as well as to others that have similar system calls. The broken
1978 code found IPv4 interfaces correctly, but gave incorrect values for the
1979 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1980 that it would not match correctly against @[] and not recognize the IPv6
1983 PH/02 The ipliteral router was not recognizing addresses of the form user@
1984 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1986 PH/03 Added disable_ipv6.
1988 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1989 decoded form, because it is most often used to construct To: headers
1990 lines in autoreplies, and the decoded form may well be syntactically
1991 invalid. However, $reply_address has leading white space removed, and all
1992 newlines turned into spaces so that the autoreply transport does not
1995 PH/05 If group was specified without a user on a router, and no group or user
1996 was specified on a transport, the group from the router was ignored.
1998 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1999 for visible compile-time settings that can be used to change these
2000 numbers, for those that want even more. Backwards compatibility with old
2001 spool files has been maintained. However, going back to a previous Exim
2002 release will lost any variables that are in spool files.
2004 PH/07 Two small changes when running in the test harness: increase delay when
2005 passing a TCP/IP connection to a new process, in case the original
2006 process has to generate a bounce, and remove special handling of
2007 127.0.0.2 (sic), which is no longer necessary.
2009 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2010 be the same on different OS.
2012 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2015 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2016 whether --show-vars was specified or not
2018 JJ/02 exipick: Added support for new ACL variable spool format introduced
2021 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2022 syntactically invalid From: or Reply-to: line, and a filter used this to
2023 generate an autoreply, and therefore failed to obtain an address for the
2024 autoreply, Exim could try to deliver to a non-existent relative file
2025 name, causing unrelated and misleading errors. What now happens is that
2026 it logs this as a hard delivery error, but does not attempt to create a
2029 PH/11 The exinext utility has a -C option for testing purposes, but although
2030 the given file was scanned by exinext itself; it wasn't being passed on
2031 when Exim was called.
2033 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2034 an end-of-file indication when reading a command response.
2036 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2037 compiled. In many other places in Exim, IPv6 addresses are always
2038 recognized, so I have changed this. It also means that IPv4 domain
2039 literals of the form [IPV4:n.n.n.n] are now always recognized.
2041 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2042 used if the router is not running as root, for example, when verifying at
2043 ACL time, or when using -bh. The debugging output from this situation was
2044 non-existent - all you got was a failure to exec. I have made two
2047 (a) Failures to set uid/gid, the current directory, or a process leader
2048 in a subprocess such as that created by queryprogram now generate
2049 suitable debugging ouput when -d is set.
2051 (b) The queryprogram router detects when it is not running as root,
2052 outputs suitable debugging information if -d is set, and then runs
2053 the subprocess without attempting to change uid/gid.
2055 PH/15 Minor change to Makefile for building test_host (undocumented testing
2058 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2059 additional section of a DNS packet that returns MX or SRV records.
2060 Instead, it always explicitly searches for A/AAAA records. This avoids
2061 major problems that occur when a DNS server includes only records of one
2062 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2063 fixed another bug: if SRV records were looked up and the corresponding
2064 address records were *not* found in the additional section, the port
2065 values from the SRV records were lost.
2067 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2068 using the correct key (the original address) when searching the retry
2069 rules in order to find which one to use for generating the retry hint.
2071 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2072 adding the default one. Similarly, if it contains a Reply-To: header, the
2073 errors_reply_to option, if set, is not used.
2075 PH/19 When calculating a retry time, Exim used to measure the "time since
2076 failure" by looking at the "first failed" field in the retry record. Now
2077 it does not use this if it is later than than the arrival time of the
2078 message. Instead it uses the arrival time. This makes for better
2079 behaviour in cases where some deliveries succeed, thus re-setting the
2080 "first failed" field. An example is a quota failure for a huge message
2081 when small messages continue to be delivered. Without this change, the
2082 "time since failure" will always be short, possible causing more frequent
2083 delivery attempts for the huge message than are intended.
2084 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2086 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2087 $1, $2, $3) because the numerical variables can be reset during some
2088 expansion items (e.g. "match"), thereby losing the authentication data.
2090 PH/21 Make -bV show the size of off_t variables so that the test suite can
2091 decide whether to run tests for quotas > 2G.
2093 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2094 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2095 filecount value is greater than 2G or if a quota value is greater than 2G
2096 on a system where the size of off_t is not greater than 4, a panic error
2099 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2100 never match. The debug and -bh output now contains an explicit error
2101 message indicating a malformed IPv4 address or mask.
2103 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2104 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2105 PH/23 above applies.
2107 PH/25 Do not write to syslog when running in the test harness. The only
2108 occasion when this arises is a failure to open the main or panic logs
2109 (for which there is an explicit test).
2111 PH/26 Added the /no_tell option to "control=freeze".
2113 PH/27 If a host name lookup failed very early in a connection, for example, if
2114 the IP address matched host_lookup and the reverse lookup yielded a name
2115 that did not have a forward lookup, an error message of the form "no IP
2116 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2117 could be logged. Now it outputs the IP address instead of "NULL".
2119 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2120 allows the sender and the authenticated sender to be set when
2121 submitting a message from within Exim. Since child_open_exim() is
2122 documented for local_scan(), the new function should be too.
2124 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2125 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2126 results in an empty string is now treated as unset.
2128 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2130 PH/31 Added sender_verify_fail logging option.
2132 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2133 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2134 tidied the source and removed it altogether.
2136 PH/33 When a queue run was abandoned because the load average was too high, a
2137 log line was always written; now it is written only if the queue_run log
2138 selector is set. In addition, the log line for abandonment now contains
2139 information about the queue run such as the pid. This is always present
2140 in "start" and "stop" lines but was omitted from the "abandon" line.
2142 PH/34 Omit spaces between a header name and the colon in the error message that
2143 is given when verify = headers_syntax fails (if there are lots of them,
2144 the message gets confusing).
2146 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2147 names, as there are now some PTR records that contain slashes. This check
2148 is only to protect against broken name servers that fall over on strange
2149 characters, so the fact that it applies to all lookups doesn't matter.
2151 PH/36 Now that the new test suite is complete, we can remove some of the
2152 special code in Exim that was needed for the old test suite. For example,
2153 sorting DNS records because real resolvers return them in an arbitrary
2154 order. The new test suite's fake resolver always returns records in the
2157 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2158 bounces) except when queue_only is set, to avoid logging races between
2159 the different processes.
2161 PH/38 Panic-die if .include specifies a non-absolute path.
2163 PH/39 A tweak to the "H" retry rule from its user.
2165 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2166 a label. They prevented compilation on older perls.
2168 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2169 a warning to be raised on newish perls.
2171 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2172 on queue. Changes to match documented behaviour of showing count of
2173 messages matching specified criteria.
2175 PH/40 Changed the default ident timeout from 30s to 5s.
2177 PH/41 Added support for the use of login_cap features, on those BSD systems
2178 that have them, for controlling the resources used by pipe deliveries.
2180 PH/42 The content-scanning code uses fopen() to create files in which to put
2181 message data. Previously it was not paying any attention to the mode of
2182 the files. Exim runs with umask(0) because the rest of the code creates
2183 files with open(), and sets the required mode explicitly. Thus, these
2184 files were ending up world-writeable. This was not a big issue, because,
2185 being within the spool directory, they were not world-accessible. I have
2186 created a function called modefopen, which takes an additional mode
2187 argument. It sets umask(777), creates the file, chmods it to the required
2188 mode, then resets the umask. All the relevant calls to fopen() in the
2189 content scanning code have been changed to use this function.
2191 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2192 to 24 hours. This avoids potential overflow problems when processing G
2193 and H retry rules. I suspect nobody ever tinkers with this value.
2195 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2197 PH/45 When the plaintext authenticator is running as a client, the server's
2198 challenges are checked to ensure they are valid base64 strings. By
2199 default, the authentication attempt is cancelled if an invalid string is
2200 received. Setting client_ignore_invalid_base64 true ignores these errors.
2201 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2202 they are received. Thus, the responses can be made to depend on the
2203 challenges. If an invalid string is ignored, an empty string is placed in
2206 PH/46 Messages that are created by the autoreply transport now contains a
2207 References: header, in accordance with RFCs 2822 and 3834.
2209 PH/47 Added authenticated_sender_force to the smtp transport.
2211 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2213 PH/49 Installed latest patch from the Sieve maintainer.
2215 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2216 was also set, the appendfile transport was unnecessarily scanning a
2217 directory of message files (e.g. for maildir delivery) to find the count
2218 of files (along with the size), even though it did not need this
2219 information. It now does the scan only if it needs to find either the
2220 size of the count of files.
2222 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2224 PH/52 Two bugs concerned with error handling when the smtp transport is
2227 (i) Exim was not creating retry information for temporary errors given
2228 for individual recipients after the DATA command when the smtp transport
2229 was used in LMTP mode. This meant that they could be retried too
2230 frequently, and not timed out correctly.
2232 (ii) Exim was setting the flag that allows error details to be returned
2233 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2234 recipients that were returned after the DATA command.
2236 PH/53 This is related to PH/52, but is more general: for any failing address,
2237 when detailed error information was permitted to be returned to the
2238 sender, but the error was temporary, then after the final timeout, only
2239 "retry timeout exceeded" was returned. Now it returns the full error as
2240 well as "retry timeout exceeded".
2242 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2243 do this, and (what is worse) MTAs that accept it.
2245 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2246 will now be deprecated.
2248 PH/56 New os.c-cygwin from the Cygwin maintainer.
2250 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2251 formats (previously only available in exim formats via -bpr, -bpru,
2252 and -bpra. Now also available in native and exiqgrep formats)
2254 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2255 with very large, slow to parse queues
2257 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2259 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2261 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2262 responses to authentication challenges, though it was showing the
2263 challenges; (ii) I've removed the CR characters from the debug output for
2266 PH/58 Allow for the insertion of a newline as well as a space when a string
2267 is turned into more than one encoded-word during RFC 2047 encoding. The
2268 Sieve code now uses this.
2270 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2271 data_4xx, lost_connection, tls_required.
2273 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2274 message was being sent as an SMTP response.
2276 PH/61 Add -l and -k options to exicyclog.
2278 PH/62 When verifying, if an address was redirected to one new address, so that
2279 verification continued, and the new address failed or deferred after
2280 having set something in $address_data, the value of $address_data was not
2281 passed back to the ACL. This was different to the case when no
2282 redirection occurred. The value is now passed back in both cases.
2284 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2285 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2286 use login_cap.h, so on its own it isn't the distinguishing feature. The
2287 new name refers directly to the setclassresources() function.
2289 PH/65 Added configuration files for NetBSD3.
2291 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2293 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2294 is preferred over IPv4.
2296 PH/68 The bounce_return_message and bounce_return_body options were not being
2297 honoured for bounces generated during the reception of non-SMTP messages.
2298 In particular, this applied to messages rejected by the ACL. This bug has
2299 been fixed. However, if bounce_return_message is true and bounce_return_
2300 body is false, the headers that are returned for a non-SMTP message
2301 include only those that have been read before the error was detected.
2302 (In the case of an ACL rejection, they have all been read.)
2304 PH/69 The HTML version of the specification is now built in a directory called
2305 spec_html instead of spec.html, because the latter looks like a path with
2306 a MIME-type, and this confuses some software.
2308 PH/70 Catch two compiler warnings in sieve.c.
2310 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2311 function verify_get_ident() calls ip_connect() to connect a socket, but
2312 if the "connect()" function timed out, ip_connect() used to close the
2313 socket. However, verify_get_ident() also closes the socket later, and in
2314 between Exim writes to the log, which may get opened at this point. When
2315 the socket was closed in ip_connect(), the log could get the same file
2316 descriptor number as the socket. This naturally causes chaos. The fix is
2317 not to close the socket in ip_connect(); the socket should be closed by
2318 the function that creates it. There was only one place in the code where
2319 this was missing, in the iplookup router, which I don't think anybody now
2320 uses, but I've fixed it anyway.
2322 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2323 well as to direct DNS lookups. Otherwise the handling of names in host
2324 lists is inconsistent and therefore confusing.
2330 PH/01 Two changes to the default runtime configuration:
2332 (1) Move the checks for relay_from_hosts and authenticated clients from
2333 after to before the (commented out) DNS black list checks.
2335 (2) Add control=submission to the relay_from_hosts and authenticated
2336 clients checks, on the grounds that messages accepted by these
2337 statements are most likely to be submissions.
2339 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2341 (1) Generate an error if the third argument for the ${prvs expansion is
2344 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2347 (3) Reset the variables that are obtained from the first argument of
2348 ${prvscheck and used in the second argument before leaving the code,
2349 because their memory is reclaimed, so using them afterwards may do
2352 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2353 one (it's much easier than Tom thought :-).
2355 (5) Because of (4), we can now allow for the use of $prvscheck_result
2356 inside the third argument.
2358 PH/03 For some reason, the default setting of PATH when running a command from
2359 a pipe transport was just "/usr/bin". I have changed it to
2362 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2363 anything to be listed in the output from -bV.
2365 PH/05 When a filter generated an autoreply, the entire To: header line was
2366 quoted in the delivery log line, like this:
2368 => >A.N.Other <ano@some.domain> <original@ddress> ...
2370 This has been changed so that it extracts the operative address. There
2371 may be more than one such address. If so, they are comma-separated, like
2374 => >ano@some.domain,ona@other.domain <original@ddress> ...
2376 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2377 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2378 not being looked up in the rDNS to get a host name, Exim was showing the
2379 IP address twice in Received: lines, even though the IP addresses were
2380 identical. For example:
2382 Received: from [1.2.3.4] (helo=[1.2.3.4])
2384 However, if the real host name was known, it was omitting the HELO data
2385 if it matched the actual IP address. This has been tidied up so that it
2386 doesn't show the same IP address twice.
2388 PH/07 When both +timestamp and +memory debugging was on, the value given by
2389 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2390 called by the debug printing, thereby overwriting the timestamp buffer.
2391 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2393 PH/08 When the original message was included in an autoreply transport, it
2394 always said "this is a copy of the message, including all the headers",
2395 even if body_only or headers_only was set. It now gives an appropriate
2398 PH/09 Applied a patch from the Sieve maintainer which:
2400 o fixes some comments
2401 o adds the (disabled) notify extension core
2402 o adds some debug output for the result of if/elsif tests
2403 o points to the current vacation draft in the documentation
2404 and documents the missing references header update
2408 o fixes a bug in processing the envelope test (when testing
2409 multiple envelope elements, the last element determinted the
2412 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2413 Electronic Mail") by including:
2415 Auto-submitted: auto-generated
2417 in the messages that it generates (bounce messages and others, such as
2418 warnings). In the case of bounce messages for non-SMTP mesages, there was
2419 also a typo: it was using "Auto_submitted" (underscore instead of
2420 hyphen). Since every message generated by Exim is necessarily in response
2421 to another message, thes have all been changed to:
2423 Auto-Submitted: auto-replied
2425 in accordance with these statements in the RFC:
2427 The auto-replied keyword:
2429 - SHOULD be used on messages sent in direct response to another
2430 message by an automatic process,
2432 - MUST NOT be used on manually-generated messages,
2434 - MAY be used on Delivery Status Notifications (DSNs) and Message
2435 Disposition Notifications (MDNs),
2437 - MUST NOT be used on messages generated by automatic or periodic
2438 processes, except for messages which are automatic responses to
2441 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2442 to the default Received: header definition.
2444 PH/12 Added log selector acl_warn_skipped (default on).
2446 PH/13 After a successful wildlsearch lookup, discard the values of numeric
2447 variables because (a) they are in the wrong storage pool and (b) even if
2448 they were copied, it wouldn't work properly because of the caching.
2450 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2451 checking when decoding. Apparently there are clients that generate
2452 overlong encoded strings. Why am I not surprised?
2454 PH/15 If the first argument of "${if match_address" was not empty, but did not
2455 contain an "@" character, Exim crashed. Now it writes a panic log message
2456 and treats the condition as false.
2458 PH/16 In autoreply, treat an empty string for "once" the same as unset.
2460 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
2461 extension "envelope-auth". The code is finished and in agreement with
2462 other implementations, but there is no documentation so far and in fact,
2463 nobody wrote the draft yet. This extension is currently #undef'ed, thus
2464 not changing the active code.
2466 Print executed "if" and "elsif" statements when debugging is used. This
2467 helps a great deal to understand what a filter does.
2469 Document more things not specified clearly in RFC3028. I had all this
2470 sorted out, when out of a sudden new issues came to my mind. Oops."
2472 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
2475 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
2476 canonical form (as documented). However, after a host name lookup from
2477 the IP address, check_host() was doing a simple string comparison with
2478 addresses acquired from the DNS when checking that the found name did
2479 have the original IP as one of its addresses. Since any found IPv6
2480 addresses are likely to be in abbreviated form, the comparison could
2481 fail. Luckily, there already exists a function for doing the comparison
2482 by converting both addresses to binary, so now that is used instead of
2483 the text comparison.
2485 PH/20 There was another similar case to PH/19, when a complete host name was
2486 given in a host list; looking up its IP address could give an abbreviated
2487 form, whereas the current host's name might or might not be abbreviated.
2488 The same fix has been applied.
2494 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
2495 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
2498 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
2499 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
2501 PH/03 Typo: missing ".o" in src/pcre/Makefile.
2503 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
2504 header line, restrict the check to what is listed in RFCs 2369 and 2929.
2505 Also, for "Auto-Submitted", treat anything other than "no" as
2506 non-personal, in accordance with RFC 3834. (Previously it treated
2507 anything starting "auto-" as non-personal.)
2509 TF/01 The control=submission/name=... option had a problem with syntax
2510 errors if the name included a slash character. The /name= option
2511 now slurps the rest of the string, so it can include any characters
2512 but it must come last in the list of options (after /sender_retain
2515 PH/05 Some modifications to the interface to the fake nameserver for the new
2523 TK/01 Added the "success_on_redirect" address verification option. See
2524 NewStuff for rationale and an example.
2526 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
2528 PH/02 Patch to exigrep to allow it to work on syslog lines.
2530 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
2531 fread() to skip over the body file's header line, because in Cygwin the
2532 header line is locked and is inaccessible.
2534 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
2535 co-exist for some time) to make it clear that it is the Exim ID that is
2536 referenced, not the Message-ID: header line.
2538 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
2539 string_format() function, because snprintf() does not exist on all
2542 PH/06 The use of forbid_filter_existstest now also locks out the use of the
2543 ${stat: expansion item.
2545 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
2546 protocol synchronization error", to keep the pedants happy.
2548 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
2549 well as for IRIX systems, when gcc is being used. See the host.c source
2552 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
2554 PH/10 Named domain lists were not working if used in a queue_smtp_domains
2557 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
2558 transport and to the smtp transport in LMTP mode.
2560 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
2562 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
2563 run a filter in a subprocess. This could lead to confusion in subsequent
2564 lookups in the parent process. There should also be a search_tidyup() at
2565 the end of the subprocess.
2567 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
2568 only if the host matched helo_try_verify_hosts, which caused the
2569 verification to occur when the EHLO/HELO command was issued. The ACL just
2570 tested the remembered result. Now, if a previous verification attempt has
2571 not happened, "verify = helo" does it there and then.
2573 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
2575 TK/03 Fix log output including CR from clamd.
2577 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
2578 exist provoked a memory error which could cause a segfault.
2580 PH/15 Installed PCRE 6.2
2582 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
2584 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
2585 of the problem. Specifically, suggested +O2 rather than +O1 for the
2588 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
2590 PH/20 If a delivery was routed to a non-standard port by means of an SRV
2591 record, the port was not correctly logged when the outgoing_port log
2592 selector was set (it logged the transort's default port).
2594 PH/21 Added support for host-specific ports to manualroute, queryprogram,
2595 fallback_hosts, and "hosts" in the smtp transport.
2597 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
2598 host errors such as "Connection refused".
2600 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
2601 authentication with radiusclient 0.4.9:
2603 - Error returned from rc_read_config was caught wrongly
2604 - Username/password not passed on to radius server due to wrong length.
2606 The presumption is that some radiusclient API changes for 4.51/PH/17
2607 were not taken care of correctly. The code is still untested by me (my
2608 Linux distribution still has 0.3.2 of radiusclient), but it was
2609 contributed by a Radius user.
2611 PH/24 When doing a callout, the value of $domain wasn't set correctly when
2612 expanding the "port" option of the smtp transport.
2614 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
2615 while reading a MIME header. Thanks to Tom Hughes for a patch.
2617 PH/24 Include config.h inside local_scan.h so that configuration settings are
2620 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
2621 that in an ACL for RCPT (for example), you can examine exactly what was
2624 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
2625 commands, but it was not correctly comparing the address with the actual
2626 client host address. Thus, it would show the EHLO address in Received:
2627 header lines when this was not necessary.
2629 PH/27 Added the % operator to ${eval:}.
2631 PH/28 Exim tries to create and chdir to its spool directory when it starts;
2632 it should be ignoring failures (because with -C, for example, it has lost
2633 privilege). It wasn't ignoring creation failures other than "already
2636 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
2639 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
2640 because an input line was too long, either on its own, or by virtue of
2641 too many continuations, the temporary file was not being removed, and the
2642 return code was incorrect.
2644 PH/31 Missing "BOOL" in function definition in filtertest.c.
2646 PH/32 Applied Sieve patches from the maintainer.
2648 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
2650 PH/33 Added "verify = not_blind".
2652 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
2653 Local/Makefile (with some defaults set). These are used in built scripts
2654 such as exicyclog, but they have never been used in the exim_install
2655 script (though there are many overriding facilities there). I have
2656 arranged that the exim_install script now takes note of these two
2659 PH/35 Installed configuration files for Dragonfly.
2661 PH/36 When a locally submitted message by a trusted user did not contain a
2662 From: header, and the sender address was obtained from -f or from an SMTP
2663 MAIL command, and the trusted user did not use -F to supply a sender
2664 name, $originator_name was incorrectly used when constructing a From:
2665 header. Furthermore, $originator_name was used for submission mode
2666 messages from external hosts without From: headers in a similar way,
2667 which is clearly wrong.
2669 PH/37 Added control=suppress_local_fixups.
2671 PH/38 When log_selector = +received_sender was set, and the addition of the
2672 sender made the log line's construction buffer exactly full, or one byte
2673 less than full, an overflow happened when the terminating "\n" was
2676 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
2677 when the result of a list match is failure because a DNS lookup failed.
2679 PH/40 RM_COMMAND is now used in the building process.
2681 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
2682 the "build-* directories that it finds.
2684 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
2685 address in a domain literal was a prefix of an interface address.
2687 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
2688 when verifying a sender address, unless rewrite_headers is false.
2690 PH/44 Wrote a long comment about why errors_to addresses are verified as
2691 recipients, not senders.
2693 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
2694 the ratelimit ACL was added.
2696 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
2698 PH/46 Added extra information about PostgreSQL errors to the error string.
2700 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
2701 avoiding the need to install special zones in a real server. This is
2702 backwards compatible; if it can't find the fake resolver, it drops back.
2703 Thus, both old and new test suites can be run.
2705 TF/02 Added util/ratelimit.pl
2707 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
2708 clock is set back in time.
2710 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
2711 Candler <B.Candler@pobox.com>.
2713 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
2714 OK for addresses that are the result of redirections.
2716 PH/48 A number of further additions for the benefit of the new test suite,
2717 including a fake gethostbyname() that interfaces to the fake DNS resolver
2720 TF/06 The fix for widen_domains has also been applied to qualify_single and
2721 search_parents which are the other dnslookup options that can cause
2724 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
2727 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
2729 TF/07 Exim produced the error message "an SRV record indicated no SMTP
2730 service" if it encountered an MX record with an empty target hostname.
2731 The message is now "an MX or SRV record indicated no SMTP service".
2733 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
2734 if the DNS of the sending site is misconfigured. This is quite a
2735 common situation. This change restores the behaviour of treating a
2736 helo verification defer as a failure.
2738 PH/51 If self=fail was set on a router, the bounce message did not include the
2739 actual error message.
2745 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
2747 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
2748 command itself ended in error, the underlying message about the transport
2749 filter timeout was being overwritten with the pipe command error. Now the
2750 underlying error message should be appended to the second error message.
2752 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
2754 PH/02 Reduce the amount of output that "make" produces by default. Full output
2755 can still be requested.
2757 PH/03 The warning log line about a condition test deferring for a "warn" verb
2758 was being output only once per connection, rather than after each
2759 occurrence (because it was using the same function as for successful
2760 "warn" verbs). This seems wrong, so I have changed it.
2762 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
2763 it should not have, which might have caused a crash in the right
2764 circumstances, but probably never did.
2766 PH/04 Installed a modified version of Tony Finch's patch to make submission
2767 mode fix the return path as well as the Sender: header line, and to
2768 add a /name= option so that you can make the user's friendly name appear
2771 TF/03 Added the control = fakedefer ACL modifier.
2773 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
2774 Mark Lowes for thorough testing.
2776 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
2778 TK/03 Merged latest SRS patch from Miles Wilton.
2780 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
2781 with the definition in sysexits.h (which is #included earlier).
2782 Fortunately, Exim does not actually use EX_OK. The code used to try to
2783 preserve the sysexits.h value, by assumimg that macro definitions were
2784 scanned for macro replacements. I have been disabused of this notion,
2785 so now the code just undefines EX_OK before #including unistd.h.
2787 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
2788 in the smtp transport. When a block could not be written in a single
2789 write() function, the timeout was being re-applied to each part-write.
2790 This seems wrong - if the receiver was accepting one byte at a time it
2791 would take for ever. The timeout is now adjusted when this happens. It
2792 doesn't have to be particularly precise.
2794 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
2795 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
2797 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
2798 without a domain if the check to <postmaster@domain> fails.
2800 SC/01 Eximstats: added -xls and the ability to specify output files
2801 (patch written by Frank Heydlauf).
2803 SC/02 Eximstats: use FileHandles for outputing results.
2805 SC/03 Eximstats: allow any combination of xls, txt, and html output.
2807 SC/04 Eximstats: fixed display of large numbers with -nvr option
2809 SC/05 Eximstats: fixed merging of reports with empty tables.
2811 SC/06 Eximstats: added the -include_original_destination flag
2813 SC/07 Eximstats: removed tabs and trailing whitespace.
2815 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
2817 TK/06 MBOX spool code: Add real "From " MBOX separator line
2818 so the .eml file is really in mbox format (even though
2819 most programs do not really care). Patch from Alex Miller.
2821 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
2822 The latter is generated from $received_to and is only set if the
2823 message has one envelope recipient. SA can use these headers,
2824 obviously out-of-the-box. Patch from Alex Miller.
2826 PH/08 The ${def test on a variable was returning false if the variable's
2827 value was "0", contrary to what the specification has always said!
2828 The result should be true unless the variable is empty.
2830 PH/09 The syntax error of a character other than { following "${if
2831 def:variable_name" (after optional whitespace) was not being diagnosed.
2832 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
2833 accidental colon was present, for example, could give incorrect results.
2835 PH/10 Tidied the code in a number of places where the st_size field of a stat()
2836 result is used (not including appendfile, where other changes are about
2839 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
2840 This involved changing a lot of size variables from int to off_t. It
2841 should work with maildirs and everything.
2843 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
2844 spamd dying while we are connected to it.
2846 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
2849 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
2850 function for BDB 4.3.
2852 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
2854 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
2855 the sources for PCRE have been re-arranged and more files are now
2858 PH/15 The code I had for printing potentially long long variables in PH/11
2859 above was not the best (it lost precision). The length of off_t variables
2860 is now inspected at build time, and an appropriate printing format (%ld
2861 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
2862 to be "long long int" or "long int". This is needed for the internal
2863 formatting function string_vformat().
2865 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
2866 the configuration file to be ":syslog", then the script "guesses" where
2867 the logs files are, rather than using the compiled in default. In our
2868 case the guess is not the same as the compiled default, so the script
2869 suddenly stopped working when I started to use syslog. The patch checks
2870 to see if log_file_path is "". If so, it attempts to read it from exim
2871 with no configuration file to get the compiled in version, before it
2872 falls back to the previous guessing code."
2874 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
2875 implementing BATV in an Exim configuration. See NewStuff for the gory
2878 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
2879 Makefile that are specific to HP-UX.
2881 PH/18 If the "use_postmaster" option was set for a recipient callout together
2882 with the "random" option, the postmaster address was used as the MAIL
2883 FROM address for the random test, but not for the subsequent recipient
2884 test. It is now used for both.
2886 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
2887 patch removes a few documentation additions to RFC 3028, because the
2888 latest draft now contains them. It adds the new en;ascii-case comparator
2889 and a new error check for 8bit text in MIME parts. Comparator and
2890 require names are now matched exactly. I enabled the subaddress
2891 extension, but it is not well tested yet (read: it works for me)."
2893 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
2894 rework some of the code of TK/09 above to avoid the hardwired use of
2895 "%lld" and "long long". Replaced the call to snprintf() with a call to
2898 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
2899 records point to non-existent hosts", "retry timeout exceeded", and
2900 "retry time not reached for any host after a long failure period".
2902 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2903 experimental DomainKeys support:
2905 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2906 (2) On an error such as an illegally used "control", the wrong name for
2907 the control was given.
2909 These problems did NOT occur unless DomainKeys support was compiled.
2911 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2913 PH/24 Added ${if match_ip condition.
2915 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2916 databases so that it will be absolutely obvious if a crash occurs in the
2917 DB library. This is a regular occurrence (often caused by mis-matched
2920 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2921 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2922 picked up on a user's system that detects such things. There doesn't seem
2923 to be a gcc warning option for this - only an attribute that has to be
2924 put on the function's prototype. It seems that in Fedora Core 4 they have
2925 set this on a number of new functions. No doubt there will be more in due
2928 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2929 specify a transport. However, if an address that was verified by such a
2930 router was the subject of a callout, Exim crashed because it tried to
2931 read the rcpt_include_affixes from the non-existent transport. Now it
2932 just assumes that the setting of that option is false. This bug was
2933 introduced by 4.51/PH/31.
2935 PH/28 Changed -d+all to exclude +memory, because that information is very
2936 rarely of interest, but it makes the output a lot bigger. People tend to
2937 do -d+all out of habit.
2939 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2940 code in os-type was giving problems when libc.so lives in lib64, like on
2943 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2944 aren't the modern standard, and it seems that some systems' include files
2945 don't always have them. Exim was already checking for some of the newer
2946 ones like T_AAAA, and defining it itself. I've added checks for all the
2947 record types that Exim uses.
2949 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2950 not automatically generating a new one, as it is supposed to. This
2951 prevented TLS from working. If the file did exist, but contained invalid
2952 data, a new version was generated, as expected. It was only the case of a
2953 non-existent file that was broken.
2955 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2956 with a change in libdomainkeys > 0.64.
2958 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2959 from DNS. If the selector record carries the flag, it now has
2960 precedence over the domain-wide flag.
2962 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2964 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2965 the use of a port name that isn't defined in /etc/services) occurred, the
2966 message was deferred as in a normal delivery, and thus remained on the
2967 spool, instead of being failed because of the mua_wrapper setting. This
2968 is now fixed, and I tidied up some of the mua_wrapper messages at the
2971 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2972 the messages in a hash of arrays rather than using individual hashes.
2973 This is a bit cleaner and results in dramatic memory savings, albeit
2974 at a slight CPU cost.
2976 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2977 as requested by Marc Sherman.
2979 SC/10 Eximstats: added histograms for user specified patterns as requested
2982 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2984 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2985 fopen() in the content-scanning modules that did not already have it.
2991 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2992 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2994 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2996 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2998 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2999 to test Sieve filters that use "vacation".
3001 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3002 that changes the way the GnuTLS parameters are stored in the cache file.
3003 The new format can be generated externally. For backward compatibility,
3004 if the data in the cache doesn't make sense, Exim assumes it has read an
3005 old-format file, and it generates new data and writes a new file. This
3006 means that you can't go back to an older release without removing the
3009 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3010 work if there are any delivery delays because "one_time" forces the
3011 parent to be marked "delivered", so its unseen clone is never tried
3012 again. For this reason, Exim now forbids the simultaneous setting of
3015 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3016 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3017 entry if you want to know the details. The fix, however, neglected to
3018 consider the case when local delivery batching is involved. The test for
3019 "previously delivered" was not happening when checking to see if an
3020 address could be batched with a previous (undelivered) one; under
3021 certain circumstances this could lead to multiple deliveries to the same
3024 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3025 in its include files, and this causes problems building Exim.
3027 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3028 header_syntax) but Exim was just ignoring anything given after a slash.
3029 In particular, this caused confusion with an attempt to use "verify =
3030 reverse_host_lookup/defer_ok". An error is now given when options are
3031 supplied for verify items that do not have them. (Maybe reverse_host_
3032 lookup should have a defer_ok option, but that's a different point.)
3034 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3035 defined by RFC 821) to 2048, because there were problems with some AUTH
3036 commands, and RFC 1869 says the size should be increased for extended
3037 SMTP commands that take arguments.
3039 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3042 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3043 "unknown" error; now it says that the functionality isn't in the binary.
3045 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3046 an address' error message when a string expansion fails (syntax or
3047 whatever). Otherwise the password may appear in the log. Following change
3048 PH/42 below, there is no longer a chance of it appearing in a bounce
3051 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3053 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3054 of its addresses was ever tried. (Bugzilla bug #2.)
3056 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3057 the result incorrectly in the debug output. (It correctly added a newline
3058 to what was transported.)
3060 TF/01 Added $received_time.
3062 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3063 commented out examples of how to interface to a virus scanner and to
3064 SpamAssassin. Also added commented examples of av_scanner and
3065 spamd_address settings.
3067 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3068 and controls are allowed in which ACLs. There were a couple of minor
3069 errors. Some of the entries in the conditions table (which is a table of
3070 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3071 negation of where the condition IS allowed.
3073 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3075 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3076 header file does not have a version number, so I've had to invent a new
3077 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3078 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3079 radiusclient), but it was contributed by a Radius user.
3081 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3082 files or directories, for OpenSSL.
3084 PH/19 When an Exim process that is running as root has to create an Exim log
3085 file, it does so in a subprocess that runs as exim:exim so as to get the
3086 ownership right at creation (otherwise, other Exim processes might see
3087 the file with the wrong ownership). There was no test for failure of this
3088 fork() call, which would lead to the process getting stuck as it waited
3089 for a non-existent subprocess. Forks do occasionally fail when resources
3090 run out. I reviewed all the other calls to fork(); they all seem to check
3093 PH/20 When checking for unexpected SMTP input at connect time (before writing
3094 the banner), Exim was not dealing correctly with a non-positive return
3095 from the read() function. If the client had disconnected by this time,
3096 the result was a log entry for a synchronization error with an empty
3097 string after "input=" when read() returned zero. If read() returned -1
3098 (an event I could not check), uninitialized data bytes were printed.
3099 There were reports of junk text (parts of files, etc) appearing after
3102 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3104 PH/22 Added support for macro redefinition, and (re)definition in between
3105 driver and ACL definitions.
3107 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3108 forgetting to use the resulting value; it was using the unexpanded value.
3110 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3111 hadn't been configured. The fix is from Juergen Kreileder, who
3112 understands it better than I do:
3114 "Here's what I see happening with three configured cyrus_sasl
3115 authenticators configured (plain, login, cram-md5):
3117 On startup auth_cyrus_sasl_init() gets called for each of these.
3118 This means three calls to sasl_listmech() without a specified mech_list.
3119 => SASL tests which mechs of all available mechs actually work
3120 => three warnings about OTP not working
3121 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3123 With the patch, sasl_listmech() also gets called three times. But now
3124 SASL's mech_list option is set to the server_mech specified in the the
3125 authenticator. Or in other words, the answer from sasl_listmech()
3126 gets limited to just the mech you're testing for (which is different
3128 => the return list contains just 'plain' or 'login', 'cram-md5' or
3129 nothing depending on the value of ob->server_mech.
3131 I've just tested the patch: Authentication still works fine,
3132 unavailable mechs specified in the exim configuration are still
3133 caught, and the auth.log warnings about OTP are gone."
3135 PH/25 When debugging is enabled, the contents of the command line are added
3136 to the debugging output, even when log_selector=+arguments is not
3139 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3140 answer is "GNU", and only if the return is "GNU/something" is the answer
3143 PH/27 $acl_verify_message is now set immediately after the failure of a
3144 verification in an ACL, and so is available in subsequent modifiers. In
3145 particular, the message can be preserved by coding like this:
3147 warn !verify = sender
3148 set acl_m0 = $acl_verify_message
3150 Previously, $acl_verify_message was set only while expanding "message"
3151 and "log_message" when a very denied access.
3153 PH/28 Modified OS/os.c-Linux with
3155 -#ifndef OS_LOAD_AVERAGE
3156 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3158 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3159 nomenclature these days.)
3161 PH/29 Installed patch from the Sieve maintainer that adds the options
3162 sieve_useraddress and sieve_subaddress to the redirect router.
3164 PH/30 In these circumstances:
3165 . Two addresses routed to the same list of hosts;
3166 . First host does not offer TLS;
3167 . First host accepts first address;
3168 . First host gives temporary error to second address;
3169 . Second host offers TLS and a TLS session is established;
3170 . Second host accepts second address.
3171 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3172 and peerdn, if requested) that were in fact used only for the second
3175 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3176 attention to any local part prefix or suffix that was matched by the
3177 router that accepted the address. It now behaves in the same way as it
3178 does for delivery: the affixes are removed from the local part unless
3179 rcpt_include_affixes is set on the transport.
3181 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3182 timeout during the DATA phase of an incoming message.
3184 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3185 applied a patch sanctioned by the Sieve maintainer.
3187 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3188 the uid or gid is negative. A case of a negative gid caused this to be
3189 noticed. The fix allows for either to be negative.
3191 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3192 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3195 PH/36 The change PH/12 above was broken. Fixed it.
3197 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3198 the grounds that routing the same address twice would always produce the
3199 same answer. This might have been true once, but it is certainly no
3200 longer true now. Routing a child address may depend on the previous
3201 routing that produced that child. Some complicated redirection strategies
3202 went wrong when messages had multiple recipients, and made Exim's
3203 behaviour dependent on the order in which the addresses were given.
3205 I have moved the duplicate checking until after the routing is complete.
3206 Exim scans the addresses that are assigned to local and remote
3207 transports, and removes any duplicates. This means that more work will be
3208 done, as duplicates will always all be routed, but duplicates are
3209 presumably rare, so I don't expect this is of any significance.
3211 For deliveries to pipes, files, and autoreplies, the duplicate checking
3212 still happens during the routing process, since they are not going to be
3215 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3216 It corrects a timeout issue with spamd. This is Ian's comment: "The
3217 background is that sometimes spamd either never reads data from a
3218 connection it has accepted, or it never writes response data. The exiscan
3219 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3220 blindly assumes that writes won't block so it may never time out."
3222 PH/39 Allow G after quota size as well as K and M.
3224 PH/40 The value set for $authenticated_id in an authenticator may not contain
3225 binary zeroes or newlines because the value is written to log lines and
3226 to spool files. There was no check on this. Now the value is run through
3227 the string_printing() function so that such characters are converted to
3228 printable escape sequences.
3230 PH/41 $message_linecount is a new variable that contains the total number of
3231 lines in the message. Compare $body_linecount, which is the count for the
3234 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3235 bounce and delay warning messages, except in certain special cases, which
3238 (a) An SMTP error message from a remote host;
3239 (b) A message specified in a :fail: redirection;
3240 (c) A message specified in a "fail" command in a system filter;
3241 (d) A message specified in a FAIL return from the queryprogram router;
3242 (e) A message specified by the cannot_route_message router option.
3244 In these cases only, Exim does include the error details in bounce and
3245 warning messages. There are also a few cases where bland messages such
3246 as "unrouteable address" or "local delivery error" are given.
3248 PH/43 $value is now also set for the "else" part of a ${run expansion.
3250 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3251 being worked on, but at least Exim now implements the latest version to
3254 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3255 process to complete was treated as a delivery failure, a timeout while
3256 writing the message to the pipe was logged, but erroneously treated as a
3257 successful delivery. Such timeouts include transport filter timeouts. For
3258 consistency with the overall process timeout, these timeouts are now
3259 treated as errors, giving rise to delivery failures by default. However,
3260 there is now a new Boolean option for the pipe transport called
3261 timeout_defer, which, if set TRUE, converts the failures into defers for
3262 both kinds of timeout. A transport filter timeout is now identified in
3265 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3266 systems where "make" and "gmake" are different, calling "gmake" at top
3267 level broke things. I've arranged for the value of $(MAKE) to be passed
3268 from the Makefile to this script so that it can call the same version of
3272 A note about Exim versions 4.44 and 4.50
3273 ----------------------------------------
3275 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3276 changes of various kinds. As a consequence, a big documentation update was
3277 needed. This delayed the release for rather longer than seemed good, especially
3278 in the light of a couple of (minor) security issues. Therefore, the changes
3279 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3280 release. So 4.44 and 4.50 are in effect two different branches that both start
3283 I have left the 4.50 change log unchanged; it contains all the changes since
3284 4.43. The change log for 4.44 is below; many of its items are identical to
3285 those for 4.50. This seems to be the most sensible way to preserve the
3286 historical information.
3292 1. Minor wording change to the doc/README.SIEVE file.
3294 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3295 computation of the current number of files was incorrect.
3297 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3298 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3301 4. Give more explanation in the error message when the command for a transport
3302 filter fails to execute.
3304 5. There are several places where Exim runs a non-Exim command in a
3305 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3306 was being done only for the command run by the queryprogram router. It is
3307 now done for all such subprocesses. The other cases are: ${run, transport
3308 filters, and the commands run by the lmtp and pipe transports.
3310 6. Added CONFIGURE_GROUP build-time option.
3312 7. Some older OS have a limit of 256 on the maximum number of file
3313 descriptors. Exim was using setrlimit() to set 1000 as a large value
3314 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3315 systems. I've change it so that if it can't get 1000, it tries for 256.
3317 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3318 was an oversight, and furthermore, ever since the addition of extra
3319 controls (e.g. 4.43/32), the checks on when to allow different forms of
3320 "control" were broken. There should now be diagnostics for all cases when a
3321 control that does not make sense is encountered.
3323 9. Added the /retain_sender option to "control=submission".
3325 10. $recipients is now available in the predata ACL (oversight).
3327 11. Tidy the search cache before the fork to do a delivery from a message
3328 received from the command line. Otherwise the child will trigger a lookup
3329 failure and thereby defer the delivery if it tries to use (for example) a
3330 cached ldap connection that the parent has called unbind on.
3332 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3333 of $address_data from the recipient verification was clobbered by the
3334 sender verification.
3336 13. The value of address_data from a sender verification is now available in
3337 $sender_address_data in subsequent conditions in the ACL statement.
3339 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3341 15. Added a new option "connect=<time>" to callout options, to set a different
3344 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3345 was its contents. (It was OK if the option was not defined at all.)
3347 17. A "Completed" log line is now written for messages that are removed from
3348 the spool by the -Mrm option.
3350 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3351 information about exactly what failed.
3353 19. Added -dd to debug only the daemon process.
3355 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3356 handles timeouts, both on the server side and network timeouts. Renamed the
3357 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3359 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3360 It is now set to "smtps".
3362 22. $host_address is now set to the target address during the checking of
3363 ignore_target_hosts.
3365 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3366 being passed; this would have caused $sender_host_name to have been used if
3367 matching the list had actually called for a host name (not very likely,
3368 since this list is usually IP addresses). A host name is now passed as
3371 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3372 code that specifies a non-restarting handler (typically sigaction() in
3373 modern systems) in an attempt to fix a rare and obscure crash bug.
3375 25. Narrowed the window for a race in the daemon that could cause it to ignore
3376 SIGCHLD signals. This is not a major problem, because they are used only to
3377 wake it up if nothing else does.
3379 26. A malformed maildirsize file could cause Exim to calculate negative values
3380 for the mailbox size or file count. Odd effects could occur as a result.
3381 The maildirsize information is now recalculated if the size or filecount
3384 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3385 support for a long time. Removed HAVE_SYS_VFS_H.
3387 28. Installed the latest version of exipick from John Jetmore.
3389 29. In an address list, if the pattern was not a regular expression, an empty
3390 subject address (from a bounce message) matched only if the pattern was an
3391 empty string. Non-empty patterns were not even tested. This was the wrong
3392 because it is perfectly reasonable to use an empty address as part of a
3393 database query. An empty address is now tested by patterns that are
3394 lookups. However, all the other forms of pattern expect the subject to
3395 contain a local part and a domain, and therefore, for them, an empty
3396 address still always fails if the pattern is not itself empty.
3398 30. Exim went into a mad DNS loop when attempting to do a callout where the
3399 host was specified on an smtp transport, and looking it up yielded more
3400 than one IP address.
3402 31. Re-factored the code for checking spool and log partition space into a
3403 function that finds that data and another that does the check. The former
3404 is then used to implement four new variables: $spool_space, $log_space,
3405 $spool_inodes, and $log_inodes.
3407 32. The RFC2047 encoding function was originally intended for short strings
3408 such as real names; it was not keeping to the 75-character limit for
3409 encoded words that the RFC imposes. It now respects the limit, and
3410 generates multiple encoded words if necessary. To be on the safe side, I
3411 have increased the buffer size for the ${rfc2047: expansion operator from
3414 33. It is now permitted to omit both strings after an "if" condition; if the
3415 condition is true, the result is "true". As before, when the second string
3416 is omitted, a false condition yields an empty string. This makes it less
3417 cumbersome to write custom ACL and router conditions.
3419 34. Failure to deliver a bounce message always caused it to be frozen, even if
3420 there was an errors_to setting on the router. The errors_to setting is now
3423 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3424 canonical form (fully expanded) before being placed in
3425 $sender_host_address.
3427 36. The table in the code that translates DNS record types into text (T_A to
3428 "A" for instance) was missing entries for NS and CNAME. It is just possible
3429 that this could have caused confusion if both these types were looked up
3430 for the same domain, because the text type is used as part of Exim's
3431 per-process caching. But the chance of anyone hitting this buglet seems
3434 37. The dnsdb lookup has been extended in a number of ways.
3436 (1) There is a new type, "zns", which walks up the domain tree until it
3437 finds some nameserver records. It should be used with care.
3439 (2) There is a new type, "mxh", which is like "mx" except that it returns
3440 just the host names, not the priorities.
3442 (3) It is now possible to give a list of domains (or IP addresses) to be
3443 looked up. The behaviour when one of the lookups defers can be
3444 controlled by a keyword.
3446 (4) It is now possible to specify the separator character for use when
3447 multiple records are returned.
3449 38. The dnslists ACL condition has been extended: it is now possible to supply
3450 a list of IP addresses and/or domains to be looked up in a particular DNS
3453 39. Added log_selector=+queue_time_overall.
3455 40. When running the queue in the test harness, wait just a tad after forking a
3456 delivery process, to get repeatability of debugging output.
3458 41. Include certificate and key file names in error message when GnuTLS fails
3459 to set them up, because the GnuTLS error message doesn't include the name
3460 of the failing file when there is a problem reading it.
3462 42. Allow both -bf and -bF in the same test run.
3464 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
3466 44. The "Exiscan patch" is now merged into the mainline Exim source.
3468 45. Sometimes the final signoff response after QUIT could fail to get
3469 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3470 < 0 before doing a fflush(). This bug looks as though it goes back to the
3471 introduction of TLS in release 3.20, but "sometimes" must have been rare
3472 because the tests only now provoked it.
3474 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
3475 (this can affect the format of dates).
3477 47. exim_tidydb, when checking for the continued existence of a message for
3478 which it has found a message-specific retry record, was not finding
3479 messages that were in split spool directories. Consequently, it was
3480 deleting retry records that should have stayed in existence.
3482 48. Steve fixed some bugs in eximstats.
3484 49. The SPA authentication driver was not abandoning authentication and moving
3485 on to the next authenticator when an expansion was forced to fail,
3486 contradicting the general specification for all authenticators. Instead it
3487 was generating a temporary error. It now behaves as specified.
3489 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
3490 (the order specifies the preference for clients). The order is now AES256,
3491 AES128, 3DES, ARCFOUR128.
3493 51. Small patch to Sieve code - explicitly set From: when generating an
3496 52. Exim crashed if a remote delivery caused a very long error message to be
3497 recorded - for instance if somebody sent an entire SpamAssassin report back
3498 as a large number of 550 error lines. This bug was coincidentally fixed by
3499 increasing the size of one of Exim's internal buffers (big_buffer) that
3500 happened as part of the Exiscan merge. However, to be on the safe side, I
3501 have made the code more robust (and fixed the comments that describe what
3504 53. Now that there can be additional text after "Completed" in log lines (if
3505 the queue_time_overall log selector is set), a one-byte patch to exigrep
3506 was needed to allow it to recognize "Completed" as not the last thing in
3509 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
3510 patch that reportedly fixes this has been added. I am not expert enough to
3511 create a test for it. This is what the patch creator wrote:
3513 "I found a little strange behaviour of ldap code when working with
3514 Windows 2003 AD Domain, where users was placed in more than one
3515 Organization Units. When I tried to give exim partial DN, the exit code
3516 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
3517 But simultaneously result of request was absolutely normal ldap result,
3518 so I produce this patch..."
3520 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
3521 REFERENCE, so I have modified the code to exclude the patch when that macro
3524 55. Some experimental protocols are using DNS PTR records for new purposes. The
3525 keys for these records are domain names, not reversed IP addresses. The
3526 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3527 leaves it alone. Component reversal etc. now happens only for IP addresses.
3530 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
3532 57. Double the size of the debug message buffer (to 2048) so that more of very
3533 long debug lines gets shown.
3535 58. The exicyclog utility now does better if the number of log files to keep
3536 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
3538 59. Two changes related to the smtp_active_hostname option:
3540 (1) $smtp_active_hostname is now available as a variable.
3541 (2) The default for smtp_banner uses $smtp_active_hostname instead
3542 of $primary_hostname.
3544 60. The host_aton() function is supposed to be passed a string that is known
3545 to be a valid IP address. However, in the case of IPv6 addresses, it was
3546 not checking this. This is a hostage to fortune. Exim now panics and dies
3547 if the condition is not met. A case was found where this could be provoked
3548 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3549 components; fortuitously, this particular loophole had already been fixed
3550 by change 4.50/55 above.
3552 If there are any other similar loopholes, the new check in host_aton()
3553 itself should stop them being exploited. The report I received stated that
3554 data on the command line could provoke the exploit when Exim was running as
3555 exim, but did not say which command line option was involved. All I could
3556 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3557 running as the user.
3560 61. There was a buffer overflow vulnerability in the SPA authentication code
3561 (which came originally from the Samba project). I have added a test to the
3562 spa_base64_to_bits() function which I hope fixes it.
3565 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
3566 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
3568 63. The daemon start-up calls getloadavg() while still root for those OS that
3569 need the first call to be done as root, but it missed one case: when
3570 deliver_queue_load_max is set with deliver_drop_privilege. This is
3571 necessary for the benefit of the queue runner, because there is no re-exec
3572 when deliver_drop_privilege is set.
3574 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
3575 This has been fixed.
3577 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3578 was in use, was not putting the data itself into the right store pool;
3579 consequently, it could be overwritten for a subsequent message in the same
3580 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3583 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
3585 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
3586 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
3587 the function were treating the return as a boolean value, which happened to
3588 work because 0=false and not-0=true, but is not correct code.
3590 68. The host_aton() function was not handling scoped IPv6 addresses (those
3591 with, for example, "%eth0" on the end) correctly.
3593 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
3594 negated items (that is, ~something) in unsigned ints. Some compilers
3595 apparently mutter when there is no cast.
3597 70. If an address verification called from an ACL failed, and did not produce a
3598 user-specific message (i.e. there was only a "system" message), nothing was
3599 put in $acl_verify_message. In this situation, it now puts the system
3602 71. Change 4.23/11 added synchronization checking at the start of an SMTP
3603 session; change 4.31/43 added the unwanted input to the log line - except
3604 that it did not do this in the start of session case. It now does.
3606 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
3607 This is wrong and can cause the other end to generate a synchronization
3608 error if it is another Exim or anything else that does the synchronization
3609 check. A QUIT command is no longer sent after a timeout.
3611 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
3612 during host lookups.
3614 74. The defer_ok option of callout verification was not working if it was used
3615 when verifying addresses in header lines, that is, for this case:
3617 verify = header_sender/callout=defer_ok
3619 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
3620 those file descriptors could be used for SMTP connections. If anything
3621 wrote to stderr (the example that came up was "warn" in embedded Perl), it
3622 could be sent to the SMTP client, causing chaos. The daemon now opens
3623 stdin, stdout, and stderr to /dev/null when it puts itself into the
3626 76. Arrange for output from Perl's "warn" command to be written to Exim's main
3627 log by default. The user can override this with suitable Perl magic.
3629 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
3630 the log message when discard triggers, was not working for the DATA ACL or
3631 for the non-SMTP ACL.
3633 78. Error message wording change in sieve.c.
3635 79. If smtp_accept_max_per_host was set, the number of connections could be
3636 restricted to fewer than expected, because the daemon was trying to set up
3637 a new connection before checking whether the processes handling previous
3638 connections had finished. The check for completed processes is now done
3639 earlier. On busy systems, this bug wouldn't be noticed because something
3640 else would have woken the daemon, and it would have reaped the completed
3643 80. If a message was submitted locally by a user whose login name contained one
3644 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
3645 It caused a spool format error. I have fixed the spool reading code. A
3646 related problem was that the "from" clause in the Received: line became
3647 illegal because of the space(s). It is now covered by ${quote_local_part.
3649 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
3652 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
3654 83. Patch from Sieve maintainer to fix unterminated string problem in
3655 "vacation" handling.
3657 84. Some minor changes to the Linux configuration files to help with other
3658 OS variants using glibc.
3660 85. One more patch for Sieve to update vacation handling to latest spec.
3663 ----------------------------------------------------
3664 See the note above about the 4.44 and 4.50 releases.
3665 ----------------------------------------------------
3671 1. Change 4.43/35 introduced a bug that caused file counts to be
3672 incorrectly computed when quota_filecount was set in an appendfile
3675 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3676 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3679 3. Give more explanation in the error message when the command for a transport
3680 filter fails to execute.
3682 4. There are several places where Exim runs a non-Exim command in a
3683 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3684 was being done only for the command run by the queryprogram router. It is
3685 now done for all such subprocesses. The other cases are: ${run, transport
3686 filters, and the commands run by the lmtp and pipe transports.
3688 5. Some older OS have a limit of 256 on the maximum number of file
3689 descriptors. Exim was using setrlimit() to set 1000 as a large value
3690 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3691 systems. I've change it so that if it can't get 1000, it tries for 256.
3693 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3694 was an oversight, and furthermore, ever since the addition of extra
3695 controls (e.g. 4.43/32), the checks on when to allow different forms of
3696 "control" were broken. There should now be diagnostics for all cases when a
3697 control that does not make sense is encountered.
3699 7. $recipients is now available in the predata ACL (oversight).
3701 8. Tidy the search cache before the fork to do a delivery from a message
3702 received from the command line. Otherwise the child will trigger a lookup
3703 failure and thereby defer the delivery if it tries to use (for example) a
3704 cached ldap connection that the parent has called unbind on.
3706 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3707 of $address_data from the recipient verification was clobbered by the
3708 sender verification.
3710 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3711 was its contents. (It was OK if the option was not defined at all.)
3713 11. A "Completed" log line is now written for messages that are removed from
3714 the spool by the -Mrm option.
3716 12. $host_address is now set to the target address during the checking of
3717 ignore_target_hosts.
3719 13. When checking ignore_target_hosts for an ipliteral router, no host name was
3720 being passed; this would have caused $sender_host_name to have been used if
3721 matching the list had actually called for a host name (not very likely,
3722 since this list is usually IP addresses). A host name is now passed as
3725 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3726 code that specifies a non-restarting handler (typically sigaction() in
3727 modern systems) in an attempt to fix a rare and obscure crash bug.
3729 15. Narrowed the window for a race in the daemon that could cause it to ignore
3730 SIGCHLD signals. This is not a major problem, because they are used only to
3731 wake it up if nothing else does.
3733 16. A malformed maildirsize file could cause Exim to calculate negative values
3734 for the mailbox size or file count. Odd effects could occur as a result.
3735 The maildirsize information is now recalculated if the size or filecount
3738 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3739 support for a long time. Removed HAVE_SYS_VFS_H.
3741 18. Updated exipick to current release from John Jetmore.
3743 19. Allow an empty sender to be matched against a lookup in an address list.
3744 Previously the only cases considered were a regular expression, or an
3747 20. Exim went into a mad DNS lookup loop when doing a callout where the
3748 host was specified on the transport, if the DNS lookup yielded more than
3751 21. The RFC2047 encoding function was originally intended for short strings
3752 such as real names; it was not keeping to the 75-character limit for
3753 encoded words that the RFC imposes. It now respects the limit, and
3754 generates multiple encoded words if necessary. To be on the safe side, I
3755 have increased the buffer size for the ${rfc2047: expansion operator from
3758 22. Failure to deliver a bounce message always caused it to be frozen, even if
3759 there was an errors_to setting on the router. The errors_to setting is now
3762 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3763 canonical form (fully expanded) before being placed in
3764 $sender_host_address.
3766 24. Updated eximstats to version 1.33
3768 25. Include certificate and key file names in error message when GnuTLS fails
3769 to set them up, because the GnuTLS error message doesn't include the name
3770 of the failing file when there is a problem reading it.
3772 26. Expand error message when OpenSSL has problems setting up cert/key files.
3775 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
3776 (this can affect the format of dates).
3778 28. exim_tidydb, when checking for the continued existence of a message for
3779 which it has found a message-specific retry record, was not finding
3780 messages that were in split spool directories. Consequently, it was
3781 deleting retry records that should have stayed in existence.
3783 29. eximstats updated to version 1.35
3784 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
3785 1.35 - bugfix such that pie charts by volume are generated correctly
3787 30. The SPA authentication driver was not abandoning authentication and moving
3788 on to the next authenticator when an expansion was forced to fail,
3789 contradicting the general specification for all authenticators. Instead it
3790 was generating a temporary error. It now behaves as specified.
3792 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
3793 (the order specifies the preference for clients). The order is now AES256,
3794 AES128, 3DES, ARCFOUR128.
3796 31. Small patch to Sieve code - explicitly set From: when generating an
3799 32. Exim crashed if a remote delivery caused a very long error message to be
3800 recorded - for instance if somebody sent an entire SpamAssassin report back
3801 as a large number of 550 error lines. This bug was coincidentally fixed by
3802 increasing the size of one of Exim's internal buffers (big_buffer) that
3803 happened as part of the Exiscan merge. However, to be on the safe side, I
3804 have made the code more robust (and fixed the comments that describe what
3807 33. Some experimental protocols are using DNS PTR records for new purposes. The
3808 keys for these records are domain names, not reversed IP addresses. The
3809 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3810 leaves it alone. Component reversal etc. now happens only for IP addresses.
3813 34. The host_aton() function is supposed to be passed a string that is known
3814 to be a valid IP address. However, in the case of IPv6 addresses, it was
3815 not checking this. This is a hostage to fortune. Exim now panics and dies
3816 if the condition is not met. A case was found where this could be provoked
3817 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3818 components; fortuitously, this particular loophole had already been fixed
3819 by change 4.50/55 or 4.44/33 above.
3821 If there are any other similar loopholes, the new check in host_aton()
3822 itself should stop them being exploited. The report I received stated that
3823 data on the command line could provoke the exploit when Exim was running as
3824 exim, but did not say which command line option was involved. All I could
3825 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3826 running as the user.
3829 35. There was a buffer overflow vulnerability in the SPA authentication code
3830 (which came originally from the Samba project). I have added a test to the
3831 spa_base64_to_bits() function which I hope fixes it.
3834 36. The daemon start-up calls getloadavg() while still root for those OS that
3835 need the first call to be done as root, but it missed one case: when
3836 deliver_queue_load_max is set with deliver_drop_privilege. This is
3837 necessary for the benefit of the queue runner, because there is no re-exec
3838 when deliver_drop_privilege is set.
3840 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3841 was in use, was not putting the data itself into the right store pool;
3842 consequently, it could be overwritten for a subsequent message in the same
3843 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3846 38. Sometimes the final signoff response after QUIT could fail to get
3847 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3848 < 0 before doing a fflush(). This bug looks as though it goes back to the
3849 introduction of TLS in release 3.20, but "sometimes" must have been rare
3850 because the tests only now provoked it.
3856 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
3857 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
3858 Now it returns an integer. A number of calls were still expecting a T/F
3859 return. Fortuitously, in all cases, the tests worked in OK situations,
3860 which is the norm. However, things would have gone wrong on any write
3861 failures on the smtp file descriptor. This function is used when sending
3862 messages over SMTP and also when doing verify callouts.
3864 2. When Exim is called to do synchronous delivery of a locally submitted
3865 message (the -odf or -odi options), it no longer closes stderr before doing
3868 3. Implemented the mua_wrapper option.
3870 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
3872 5. Implemented the functions header_remove(), header_testname(),
3873 header_add_at_position(), and receive_remove_recipient(), and exported them
3876 6. If an ACL "warn" statement specified the addition of headers, Exim already
3877 inserted X-ACL-Warn: at the start if there was no header name. However, it
3878 was not making this test for the second and subsequent header lines if
3879 there were newlines in the string. This meant that an invalid header could
3880 be inserted if Exim was badly configured.
3882 7. Allow an ACL "warn" statement to add header lines at the start or after all
3883 the Received: headers, as well as at the end.
3885 8. Added the rcpt_4xx retry error code.
3887 9. Added postmaster_mailfrom=xxx to callout verification option.
3889 10. Added mailfrom=xxxx to the callout verification option, for verify=
3892 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
3893 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
3895 12. Inserted some casts to stop certain compilers warning when using pointer
3896 differences as field lengths or precisions in printf-type calls (mostly
3897 affecting debugging statements).
3899 13. Added optional readline() support for -be (dynamically loaded).
3901 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
3902 same clock tick as a message's arrival, so that its received time was the
3903 same as the "first fail" time on the retry record, and that message
3904 remained on the queue past the ultimate address timeout, every queue runner
3905 would try a delivery (because it was past the ultimate address timeout) but
3906 after another failure, the ultimate address timeout, which should have then
3907 bounced the address, did not kick in. This was a "< instead of <=" error;
3908 in most cases the first failure would have been in the next clock tick
3909 after the received time, and all would be well.
3911 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
3912 being recognized when the domain list was tested by the match_domain
3913 condition in an expansion string.
3915 16. Added the ${str2b64: operator.
3917 17. Exim was always calling setrlimit() to set a large limit for the number of
3918 processes, without checking whether the existing limit was already
3919 adequate. (It did check for the limit on file descriptors.) Furthermore,
3920 errors from getrlimit() and setrlimit() were being ignored. Now they are
3921 logged to the main and panic logs, but Exim does carry on, to try to do its
3922 job under whatever limits there are.
3924 18. Imported PCRE 5.0.
3926 19. Trivial typo in log message " temporarily refused connection" (the leading
3929 20. If the log selector return_path_on_delivery was set and an address was
3930 redirected to /dev/null, the delivery process crashed because it assumed
3931 that a return path would always be set for a "successful" delivery. In this
3932 case, the whole delivery is bypassed as an optimization, and therefore no
3935 21. Internal re-arrangement: the function for sending a challenge and reading
3936 a response while authentication was assuming a zero-terminated challenge
3937 string. It's now changed to take a pointer and a length, to allow for
3938 binary data in such strings.
3940 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3942 23. Exim was not respecting finduser_retries when seeking the login of the
3943 uid under which it was called; it was always trying 10 times. (The default
3944 setting of finduser_retries is zero.) Also, it was sleeping after the final
3945 failure, which is pointless.
3947 24. Implemented tls_on_connect_ports.
3949 25. Implemented acl_smtp_predata.
3951 26. If the domain in control=submission is set empty, Exim assumes that the
3952 authenticated id is a complete email address when it generates From: or
3953 Sender: header lines.
3955 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3956 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3957 chown and chgrp in /bin and hostname in /usr/bin.
3959 28. Exim was keeping the "process log" file open after each use, just as it
3960 does for the main log. This opens the possibility of it remaining open for
3961 long periods when the USR1 signal hits a daemon. Occasional processlog
3962 errors were reported, that could have been caused by this. Anyway, it seems
3963 much more sensible not to leave this file open at all, so that is what now
3966 29. The long-running daemon process does not normally write to the log once it
3967 has entered its main loop, and it closes the log before doing so. This is
3968 so that log files can straightforwardly be renamed and moved. However,
3969 there are a couple of unusual error situations where the daemon does write
3970 log entries, and I had neglected to close the log afterwards.
3972 30. The text of an SMTP error response that was received during a remote
3973 delivery was being truncated at 512 bytes. This is too short for some of
3974 the long messages that one sometimes sees. I've increased the limit to
3977 31. It is now possible to make retry rules that apply only when a message has a
3978 specific sender, in particular, an empty sender.
3980 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3981 it possible to be selective about when SMTP synchronization is enforced.
3983 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3985 32. Implemented hosts_connection_nolog.
3987 33. Added an ACL for QUIT.
3989 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3992 35. Added mailbox_size and mailbox_filecount to appendfile.
3994 36. Added control = no_multiline_responses to ACLs.
3996 37. There was a bug in the logic of the code that waits for the clock to tick
3997 in the case where the clock went backwards by a substantial amount such
3998 that the microsecond fraction of "now" was more than the microsecond
3999 fraction of "then" (but the whole seconds number was less).
4001 38. Added support for the libradius Radius client library this is found on
4002 FreeBSD (previously only the radiusclient library was supported).
4008 1. When certain lookups returned multiple values in the form name=value, the
4009 quoting of the values was not always being done properly. Specifically:
4010 (a) If the value started with a double quote, but contained no whitespace,
4012 (b) If the value contained whitespace other than a space character (i.e.
4013 tabs or newlines or carriage returns) it was not quoted.
4014 This fix has been applied to the mysql and pgsql lookups by writing a
4015 separate quoting function and calling it from the lookup code. The fix
4016 should probably also be applied to nisplus, ibase and oracle lookups, but
4017 since I cannot test any of those, I have not disturbed their existing code.
4019 2. A hit in the callout cache for a specific address caused a log line with no
4020 reason for rejecting RCPT. Now it says "Previous (cached) callout
4021 verification failure".
4023 3. There was an off-by-one bug in the queryprogram router. An over-long
4024 return line was truncated at 256 instead of 255 characters, thereby
4025 overflowing its buffer with the terminating zero. As well as fixing this, I
4026 have increased the buffer size to 1024 (and made a note to document this).
4028 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4029 when Exim is waiting for an SMTP response from a remote server, Exim
4030 restarts its select() call on the socket, thereby resetting its timeout.
4031 This is not a problem when such interrupts are rare. Somebody set up a cron
4032 job to run exiwhat every 2 minutes, which is less than the normal select()
4033 timeout (5 or 10 minutes). This meant that the select() timeout never
4034 kicked in because it was always reset. I have fixed this by comparing the
4035 time when an interrupt arrives with the time at the start of the first call
4036 to select(). If more time than the timeout has elapsed, the interrupt is
4037 treated as a timeout.
4039 5. Some internal re-factoring in preparation for the addition of Sieve
4040 extensions (by MH). In particular, the "personal" test is moved to a
4041 separate function, and given an option for scanning Cc: and Bcc: (which is
4042 not set for Exim filters).
4044 6. When Exim created an email address using the login of the caller as the
4045 local part (e.g. when creating a From: or Sender: header line), it was not
4046 quoting the local part when it contained special characters such as @.
4048 7. Installed new OpenBSD configuration files.
4050 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4051 try to make them clearer.
4053 9. Callout options, other than the timeout value, were being ignored when
4054 verifying sender addresses in header lines. For example, when using
4056 verify = header_sender/callout=no_cache
4058 the cache was (incorrectly) being used.
4060 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4061 only the Cygwin environment.
4063 11. When return_path_on_delivery was set as a log selector, if different remote
4064 addresses in the same message used different return paths and parallel
4065 remote delivery occurred, the wrong values would sometimes be logged.
4066 (Whenever a remote delivery process finished, the return path value from
4067 the most recently started remote delivery process was logged.)
4069 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4070 lines when AUTH and/or TLS are in use. This is the "received protocol"
4071 field. Exim used to use "asmtp" for authenticated SMTP, without any
4072 indication (in the protocol name) for TLS use. Now it follows the RFC and
4073 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4074 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4075 names appear in log lines as well as in Received: header lines.
4077 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4078 extensions, and comparison tests, and to fix some bugs.
4080 14. Changes to the "personal" filter test:
4082 (1) The test was buggy in that it was just doing the equivalent of
4083 "contains" tests on header lines. For example, if a user's address was
4084 anne@some.where, the "personal" test would incorrectly be true for
4086 To: susanne@some.where
4088 This test is now done by extracting each address from the header in turn,
4089 and checking the entire address. Other tests that are part of "personal"
4090 are now done using regular expressions (for example, to check local parts
4091 of addresses in From: header lines).
4093 (2) The list of non-personal local parts in From: addresses has been
4094 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4095 taken from the Sieve specification recommendations.
4097 (3) If the message contains any header line starting with "List-" it is
4098 treated as non-personal.
4100 (4) The test for "circular" in the Subject: header line has been removed
4101 because it now seems ill-conceived.
4103 15. Minor typos in src/EDITME comments corrected.
4105 16. Installed latest exipick from John Jetmore.
4107 17. If headers_add on a router specified a text string that was too long for
4108 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4109 of string_sprintf() is now avoided.
4111 18. $message_body_size was not set (it was always zero) when running the DATA
4112 ACL and the local_scan() function.
4114 19. For the "mail" command in an Exim filter, no default was being set for
4115 the once_repeat time, causing a random time value to be used if "once" was
4116 specified. (If the value happened to be <= 0, no repeat happened.) The
4117 default is now 0s, meaning "never repeat". The "vacation" command was OK
4118 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4119 (I found it when inspecting the code).
4121 20. There is now an overall timeout for performing a callout verification. It
4122 defaults to 4 times the callout timeout, which applies to individual SMTP
4123 commands during the callout. The overall timeout applies when there is more
4124 than one host that can be tried. The timeout is checked before trying the
4125 next host. This prevents very long delays if there are a large number of
4126 hosts and all are timing out (e.g. when the network connections are timing
4127 out). The value of the overall timeout can be changed by specifying an
4128 additional sub-option for "callout", called "maxwait". For example:
4130 verify = sender/callout=5s,maxwait=20s
4132 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4133 to the end before writing, but this should make it even safer).
4135 22. Exim was forgetting that it had advertised PIPELINING for the second and
4136 subsequent messages on an SMTP connection. It was also not resetting its
4137 memory on STARTTLS and an internal HELO.
4139 23. When Exim logs an SMTP synchronization error within a session, it now
4140 records whether PIPELINING has been advertised or not.
4142 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4143 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4144 rather than long int.
4146 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4148 26. Added the never_mail option to autoreply.
4154 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4155 crash if the getsockname() call failed; this can happen if a connection is
4156 closed very soon after it is established. The problem was simply in the
4157 order in which certain operations were done, causing Exim to try to write
4158 to the SMTP stream before it had set up the file descriptor. The bug has
4159 been fixed by making things happen in the correct order.
4165 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4166 before the connection was closed, thus losing the rejection response.
4168 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4169 some early Solaris releases, but causes trouble in current releases where
4170 socklen_t is defined.
4172 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4175 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4178 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4179 that releasing the top of it at the end releases what was used for sub-
4180 expansions (unless the block got too big). However, discard this block if
4181 the first thing is a variable or header, so that we can use its block when
4182 it is dynamic (useful for very large $message_headers, for example).
4184 6. Lookups now cache *every* query, not just the most recent. A new, separate
4185 store pool is used for this. It can be recovered when all lookup caches are
4186 flushed. Lookups now release memory at the end of their result strings.
4187 This has involved some general refactoring of the lookup sources.
4189 7. Some code has been added to the store_xxx() functions to reduce the amount
4190 of flapping under certain conditions.
4192 8. log_incoming_interface used to affect only the <= reception log lines. Now
4193 it causes the local interface and port to be added to several more SMTP log
4194 lines, for example "SMTP connection from", and rejection lines.
4196 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4198 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4200 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4201 could be overwritten at the end of the current message (or the start of a
4202 new message if it was set in a HELO ACL). The value is now preserved for
4203 the duration of the SMTP connection.
4205 12. If a transport had a headers_rewrite setting, and a matching header line
4206 contained an unqualified address, that address was qualified, even if it
4207 did not match any rewriting rules. The underlying bug was that the values
4208 of the flags that permit the existence of unqualified sender and recipient
4209 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4210 non-local messages, and by -bnq for local messages) were not being
4211 preserved with the message after it was received.
4213 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4214 "next input=" as part of the text comprising the host identity instead of
4215 the correct text. The code was using the same buffer for two different
4216 strings. However, depending on which order the printing function evaluated
4217 its arguments, the bug did not always show up. Under Linux, for example, my
4218 test suite worked just fine.
4220 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4221 doesn't work with some older versions of Perl. It has been changed to "my",
4222 which in any case is probably the better facility to use.
4224 15. A really picky compiler found some instances of statements for creating
4225 error messages that either had too many or two few arguments for the format
4228 16. The size of the buffer for calls to the DNS resolver has been increased
4229 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4230 for addresses that have a lot of PTR records. This alleviates a problem; it
4231 does not fully solve it.
4233 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4234 buffer now truncates the list and logs the incident, which is the same
4235 action as happens when Exim is looking up a host name and its aliases.
4236 Previously in this situation something unpredictable would happen;
4237 sometimes it was "internal error: store_reset failed".
4239 18. If a server dropped the connection unexpectedly when an Exim client was
4240 using GnuTLS and trying to read a response, the client delivery process
4241 crashed while trying to generate an error log message.
4243 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4244 string, for example:
4246 warn message = H1: something\nH2: something
4248 the text was added as a single header line from Exim's point of view
4249 though it ended up OK in the delivered message. However, searching for the
4250 second and subsequent header lines using $h_h2: did not work. This has been
4251 fixed. Similarly, if a system filter added multiple headers in this way,
4252 the routers could not see them.
4254 20. Expanded the error message when iplsearch is called with an invalid key to
4255 suggest using net-iplsearch in a host list.
4257 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4258 ACLs are no longer actually imposed (and a message to that effect is
4261 22. If a "gecos" field in a passwd entry contained escaped characters, in
4262 particular, if it contained a \" sequence, Exim got it wrong when building
4263 a From: or a Sender: header from that name. A second bug also caused
4264 incorrect handling when an unquoted " was present following a character
4265 that needed quoting.
4267 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4268 was not being matched caselessly.
4270 24. Arranged for all hyphens in the exim.8 source to be escaped with
4273 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4274 properly. Recipient callout cache records were still being keyed to include
4275 the sender, even when use_sender was set false. This led to far more
4276 callouts that were necessary. The sender is no longer included in the key
4277 when use_sender is false.
4279 26. Added "control = submission" modifier to ACLs.
4281 27. Added the ${base62d: operator to decode base 62 numbers.
4283 28. dnsdb lookups can now access SRV records.
4285 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4286 the configuration file.
4288 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4289 (-v) mode. This makes the output for a verbose queue run more intelligible.
4291 31. Added a use_postmaster feature to recipient callouts.
4293 32. Added the $body_zerocount variable, containing the number of binary zero
4294 bytes in the message body.
4296 33. The time of last modification of the "new" subdirectory is now used as the
4297 "mailbox time last read" when there is a quota error for a maildir
4300 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4302 35. Added +ignore_unknown as a special item in host lists.
4304 36. Code for decoding IPv6 addresses in host lists is now included, even if
4305 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4306 address was recognized as an IP address, but was then not correctly decoded
4307 into binary, causing unexpected and incorrect effects when compared with
4314 1. Very minor rewording of debugging text in manualroute to say "list of
4315 hosts" instead of "hostlist".
4317 2. If verify=header_syntax was set, and a header line with an unqualified
4318 address (no domain) and a large number of spaces between the end of the
4319 name and the colon was received, the reception process suffered a buffer
4320 overflow, and (when I tested it) crashed. This was caused by some obsolete
4321 code that should have been removed. The fix is to remove it!
4323 3. When running in the test harness, delay a bit after writing a bounce
4324 message to get a bit more predictability in the log output.
4326 4. Added a call to search_tidyup() just before forking a reception process. In
4327 theory, someone could use a lookup in the expansion of smtp_accept_max_
4328 per_host which, without the tidyup, could leave open a database connection.
4330 5. Added the variables $recipient_data and $sender_data which get set from a
4331 lookup success in an ACL "recipients" or "senders" condition, or a router
4332 "senders" option, similar to $domain_data and $local_part_data.
4334 6. Moved the writing of debug_print from before to after the "senders" test
4337 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4338 problems for message scanning, either using a data ACL, or using
4339 local_scan() because the Received: header was not generated till after they
4340 were called (in order to set the time as the time of reception completion).
4341 I have revised the way this works. The header is now generated after the
4342 body is received, but before the ACL or local_scan() are called. After they
4343 are run, the timestamp in the header is updated.
4349 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4350 before starting a queue runner without re-exec. This happened only when
4351 deliver_drop_privilege was set or when the Exim user was set to root. The
4352 effect of the bug was that timeouts during subsequent deliveries caused
4353 crashes instead of being properly handled. The handler is now left at its
4354 default (and expected) setting.
4356 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4357 message, again when deliver_drop_privilege is set or Exim is run as root.
4358 The bug described in (1) was not present in this case, but the tidying up
4359 of the other signals was missing. I have made the two cases consistent.
4361 3. The ignore_target_hosts setting on a manualroute router was being ignored
4362 for hosts that were looked up using the /MX notation.
4364 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4367 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4368 operated on the sender address. After changing the $sender_address to <>
4369 for the sender address verify, Exim was re-instated it as the original
4370 (before rewriting) address, but remembering that it had rewritten it, so it
4371 wasn't rewriting it again. This bug also had the effect of breaking the
4372 sender address verification caching when the sender address was rewritten.
4374 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4375 This has been changed so that if the ip literal address matches
4376 ignore_target_hosts, the router declines.
4378 7. Added expansion conditions match_domain, match_address, and match_local_
4379 part (NOT match_host).
4381 8. The placeholder for the Received: header didn't have a length field set.
4383 9. Added code to Exim itself and to exim_lock to test for a specific race
4384 condition that could lead to file corruption when using MBX delivery. The
4385 issue is with the lockfile that is created in /tmp. If this file is removed
4386 after a process has opened it but before that process has acquired a lock,
4387 there is the potential for a second process to recreate the file and also
4388 acquire a lock. This could lead to two Exim processes writing to the file
4389 at the same time. The added code performs the same test as UW imapd; it
4390 checks after acquiring the lock that its file descriptor still refers to
4391 the same named file.
4393 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4394 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4397 11. Added the smtp_active_hostname option. If used, this will typically be made
4398 to depend on the incoming interface address. Because $interface_address is
4399 not set up until the daemon has forked a reception process, error responses
4400 that can happen earlier (such as "too many connections") no longer contain
4403 12. If an expansion in a condition on a "warn" statement fails because a lookup
4404 defers, the "warn" statement is abandoned, and the next ACL statement is
4405 processed. Previously this caused the whole ACL to be aborted.
4407 13. Added the iplsearch lookup type.
4409 14. Added ident_timeout as a log selector.
4411 15. Added tls_certificate_verified as a log selector.
4413 16. Added a global option tls_require_ciphers (compare the smtp transport
4414 option of the same name). This controls incoming TLS connections.
4416 17. I finally figured out how to make tls_require_ciphers do a similar thing
4417 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
4418 before starting the TLS session.
4420 18. Tabs are now shown as \t in -bP output.
4422 19. If the log selector return_path_on_delivery was set, Exim crashed when
4423 bouncing a message because it had too many Received: header lines.
4425 20. If two routers both had headers_remove settings, and the first one included
4426 a superfluous trailing colon, the final name in the first list and the
4427 first name in the second list were incorrectly joined into one item (with a
4428 colon in the middle).
4434 1. Added -C and -D options to the exinext utility, mainly to make it easier
4435 to include in the automated testing, but these could be helpful when
4436 multiple configurations are in use.
4438 2. The exinext utility was not formatting the output nicely when there was
4439 an alternate port involved in the retry record key, nor when there was a
4440 message id as well (for retries that were specific to a specific message
4441 and a specific host). It was also confused by IPv6 addresses, because of
4442 the additional colons they contain. I have fixed the IPv4 problem, and
4443 patched it up to do a reasonable job for IPv6.
4445 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
4446 delivery, the log line now contains "pipelined" if PIPELINING was used.
4448 4. An SMTP transport process used to panic and die if the bind() call to set
4449 an explicit outgoing interface failed. This has been changed; it is now
4450 treated in the same way as a connect() failure.
4452 5. A reference to $sender_host_name in the part of a conditional expansion
4453 that was being skipped was still causing a DNS lookup. This no longer
4456 6. The def: expansion condition was not recognizing references to header lines
4457 that used bh_ and bheader_.
4459 7. Added the _cache feature to named lists.
4461 8. The code for checking quota_filecount in the appendfile transport was
4462 allowing one more file than it should have been.
4464 9. For compatibility with Sendmail, the command line option
4472 and sets the incoming protocol and host name (for trusted callers). The
4473 host name and its colon can be omitted when only the protocol is to be set.
4474 Note the Exim already has two private options, -pd and -ps, that refer to
4475 embedded Perl. It is therefore impossible to set a protocol value of "d" or
4476 "s", but I don't think that's a major issue.
4478 10. A number of refactoring changes to the code, none of which should affect
4481 (a) The number of logging options was getting close to filling up the
4482 32-bit word that was used as a bit map. I have split them into two classes:
4483 those that are passed in the argument to log_write(), and those that are
4484 only ever tested independently outside of that function. These are now in
4485 separate 32-bit words, so there is plenty of room for expansion again.
4486 There is no change in the user interface or the logging behaviour.
4488 (b) When building, for example, log lines, the code previously used a
4489 macro that called string_cat() twice, in order to add two strings. This is
4490 not really sufficiently general. Furthermore, there was one instance where
4491 it was actually wrong because one of the argument was used twice, and in
4492 one call a function was used. (As it happened, calling the function twice
4493 did not affect the overall behaviour.) The macro has been replaced by a
4494 function that can join an arbitrary number of extra strings onto a growing
4497 (c) The code for expansion conditions now uses a table and a binary chop
4498 instead of a serial search (which was left over from when there were very
4499 few conditions). Also, it now recognizes conditions like "pam" even when
4500 the relevant support is not compiled in: a suitably worded error message is
4501 given if an attempt is made to use such a condition.
4503 11. Added ${time_interval:xxxxx}.
4505 12. A bug was causing one of the ddress fields not to be passed back correctly
4506 from remote delivery subprocesses. The field in question was not being
4507 subsequently used, so this caused to problems in practice.
4509 13. Added new log selectors queue_time and deliver_time.
4511 14. Might have fixed a bug in maildirsizefile handling that threw up
4512 "unexpected character" debug warnings, and recalculated the data
4513 unnecessarily. In any case, I expanded the warning message to give more
4516 15. Added the message "Restricted characters in address" to the statements in
4517 the default ACL that block characters like @ and % in local parts.
4519 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
4520 Three changes have been made:
4522 (a) There was a serious bug; a negative response to MAIL caused the whole
4523 recipient domain to be cached as invalid, thereby blocking all messages
4524 to all local parts at the same domain, from all senders. This bug has
4525 been fixed. The domain is no longer cached after a negative response to
4526 MAIL if the sender used is not empty.
4528 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
4531 (c) A new callout option, "use_sender" has been added for people who want
4532 the modified behaviour.
4538 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
4541 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
4542 indeed breaks things for older releases.
4544 3. Added additional logging to the case where there is a problem reading data
4545 from a filter that is running in a subprocess using a pipe, in order to
4546 try to track down a specific problem.
4548 4. Testing facility fudge: when running in the test harness and attempting
4549 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
4550 getting "No route to host". Convert this to a timeout.
4552 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
4555 6. Some OS don't have socklen_t but use size_t instead. This affects the
4556 fifth argument of getsockopt() amongst other things. This is now
4557 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
4558 can be set for individual OS. I have set it for SunOS5, OSF1, and
4559 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
4560 some earlier ones do not.
4562 7. Change 4.30/15 was not doing the test caselessly.
4564 8. The standard form for an IPv6 address literal was being rejected by address
4565 parsing in, for example, MAIL and RCPT commands. An example of this kind of
4566 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4567 this, as well as the form without the "IPv6" on the front (but only when
4568 address literals are enabled, of course).
4570 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
4572 10. Exim crashed if a message with an empty sender address specified by -f
4573 encountered a router with an errors_to setting. This could be provoked only
4574 by a command such as
4578 where an empty string was supplied; "<>" did not hit this bug.
4580 11. Installed PCRE release 4.5.
4582 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
4583 remained set. It is now erased.
4585 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
4586 times from message ids (which are base 36 rather than the normal 62).
4588 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
4589 were being counted as actual protocol errors, and logged if the log
4590 selector +smtp_protocol_error was set. One cannot be perfect in this test,
4591 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
4592 and DATA following a set of rejected RCPTs do not count as protocol errors.
4593 In other words, Exim assumes they were pipelined, though this may not
4594 actually be the case. Of course, in all cases the client gets an
4595 appropriate error code.
4597 15. If a lookup fails in an ACL condition, a message about the failure may
4598 be available; it is used if testing the ACL cannot continue, because most
4599 such messages specify what the cause of the deferral is. However, some
4600 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
4601 that caused an old message to be retained and used if a later statement
4602 caused a defer, replacing the real cause of the deferral.
4604 16. If an IP address had so many PTR records that the DNS lookup buffer
4605 was not large enough to hold them, Exim could crash while trying to process
4606 the truncated data. It now detects and logs this case.
4608 17. Further to 4.21/58, another change has been made: if (and only if) the
4609 first line of a message (the first header line) ends with CRLF, a bare LF
4610 in a subsequent header line has a space inserted after it, so as not to
4611 terminate the header.
4613 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
4614 unnecessarily, used atoi() instead of strtol(), and didn't check the
4615 termination when getting file sizes from file names by regex.
4617 19. Completely re-implemented the support for maildirsize files, in the light
4618 of a number of problems with the previous contributed implementation
4619 (4.30/29). In particular:
4621 . If the quota is zero, the maildirsize file is maintained, but no quota is
4624 . If the maildir directory does not exist, it is created before any attempt
4625 to write a maildirsize file.
4627 . The quota value in the file is just a cache; if the quota is changed in
4628 the transport, the new value overrides.
4630 . A regular expression is available for excluding directories from the
4633 20. The autoreply transport checks the characters in options that define the
4634 message's headers; it allows continued headers, but it was checking with
4635 isspace() after an embedded newline instead of explicitly looking for a
4638 21. If all the "regular" hosts to which an address was routed had passed their
4639 expiry times, and had not reached their retry times, the address was
4640 bounced, even if fallback hosts were defined. Now Exim should go on to try
4643 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
4644 equivalent code in the SMTP transport. Some hosts send humungous responses
4645 to HELO/EHLO, more than 1024 it seems.
4647 23. Refactoring: code in filter.c used (void *) for "any old type" but this
4648 gives compiler warnings in some environments. I've now done it "properly",
4651 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
4652 (because of problems with the built-in one) was declared to return uschar *
4653 instead of char *, causing compiler failure.
4655 25. Fixed a file descriptor leak when processing alias/forward files.
4657 26. Fixed a minor format string issue in dbfn.c.
4659 27. Typo in exim.c: ("dmbnz" for "dbmnz").
4661 28. If a filter file refered to $h_xxx or $message_headers, and the headers
4662 contained RFC 2047 "words", Exim's memory could, under certain conditions,
4665 29. When a sender address is verified, it is cached, to save repeating the test
4666 when there is more than one recipient in a message. However, when the
4667 verification involves a callout, it is possible for different callout
4668 options to be set for different recipients. It is too complicated to keep
4669 track of this in the cache, so now Exim always runs a verification when a
4670 callout is required, relying on the callout cache for the optimization.
4671 The overhead is duplication of the address routing, but this should not be
4674 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
4675 to be verified with callout=postmaster, and the main callout worked but the
4676 postmaster check failed, the verification correctly failed. However, if a
4677 subsequent RCPT command asked for sender verification *without* the
4678 postmaster check, incorrect caching caused this verification also to fail,
4681 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
4682 it was not caching the DNS options (qualify_single, search_parents) that
4683 were used when the lookup failed. A subsequent lookup with different
4684 options therefore always gave the same answer, though there were cases
4685 where it should not have. (Example: a "domains = !$mx_any" option on a
4686 dnslookup router: the "domains" option is always processed without any
4687 widening, but the router might have qualify_single set.) Now Exim uses the
4688 cached value only when the same options are set.
4690 32. Added John Jetmore's "exipick" utility to the distribution.
4692 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
4693 than a timeout (e.g. a certificate is required, and is not provided), an
4694 Exim server now closes the connection immediately. Previously it waited for
4695 the client to close - but if the client is SSL, it seems that they each
4696 wait for each other, leading to a delay before one of them times out.
4698 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
4699 maintained 0.8.x compatibility because I don't think many are using it, and
4700 it is clearly obsolete.
4702 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
4705 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
4706 client certificate was expired. A simple patch fixes this, though I don't
4707 understand the full logic of why the verify callback is called multiple
4710 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
4711 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
4712 which causes problems with some clients (such as the Certicom SSL Plus
4713 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
4714 disables the coutermeasure allowing Eudora to connect."
4716 38. Exim was not checking that a write() to a log file succeeded. This could
4717 lead to Bad Things if a log got too big, in particular if it hit a file
4718 size limit. Exim now panics and dies if it cannot write to a log file, just
4719 as it does if it cannot open a log file.
4721 39. Modified OS/Makefile-Linux so that it now contains
4723 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
4725 The two -D definitions ensure that Exim is compiled with large file
4726 support, which makes it possible to handle log files that are bigger than
4729 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
4730 instance) a domain was checked against a named list that involved a lookup,
4731 causing $domain_data to be set, then another domain was checked against the
4732 same list, then the first domain was re-checked, the value of $domain_data
4733 after the final check could be wrong. In particular, if the second check
4734 failed, it could be set empty. This bug probably also applied to
4737 41. The strip_trailing_dot option was not being applied to the address given
4738 with the -f command-line option.
4740 42. The code for reading a message's header from the spool was incrementing
4741 $received_count, but never initializing it. This meant that the value was
4742 incorrect (doubled) while delivering a message in the same process in which
4743 it was received. In the most common configuration of Exim, this never
4744 happens - a fresh exec is done - but it can happen when
4745 deliver_drop_privilege is set.
4747 43. When Exim logs an SMTP synchronization error - client data sent too soon -
4748 it now includes up to 150 characters of the unexpected data in the log
4751 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
4752 and building data strings. The size of both of these buffers was 10 000
4753 bytes - far larger than anybody would *ever* want, thought I. Needless to
4754 say, somebody hit the limit. I have increased the maximum line length to
4755 20 000 and the maximum data length of concatenated lines to 100 000. I have
4756 also fixed two bugs, because there was no checking on these buffers. Tsk,
4757 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
4758 buffer is too small.
4760 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
4761 lsearch lookups. Now it does.
4763 46. When parsing a route_list item in a manualroute router, a fixed-length
4764 buffer was used for the list of hosts. I made this 1024 bytes long,
4765 thinking that nobody would ever have a list of hosts that long. Wrong.
4766 Somebody had a whole pile of complicated expansion conditions, and the
4767 string was silently truncated, leading to an expansion error. It turns out
4768 that it is easier to change to an unlimited length (owing to other changes
4769 that have happened since this code was originally written) than to build
4770 structure for giving a limitation error. The length of the item that
4771 expands into the list of hosts is now unlimited.
4773 47. The lsearch lookup could not handle data where the length of text line was
4774 more than 4095 characters. Such lines were truncated, leading to shortened
4775 data being returned. It should now handle lines of any length.
4777 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
4778 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
4781 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
4782 replacing strings such as BIN_DIRECTORY in a source file: the replacement
4783 no longer happens in comment lines. A list of replacements is now placed
4784 at the head of all of the source files, except those whose only change is
4785 to replace PERL_COMMAND in the very first #! line.
4787 50. Replaced the slow insertion sort in queue.c, for sorting the list of
4788 messages on the queue, with a bottom-up merge sort, using code contributed
4789 by Michael Haardt. This should make operations like -bp somewhat faster on
4790 large queues. It won't affect queue runners, except when queue_run_in_order
4793 51. Installed eximstats 1.31 in the distribution.
4795 52. Added support for SRV lookups to the dnslookup router.
4797 53. If an ACL referred to $message_body or $message_body_end, the value was not
4798 reset for any messages that followed in the same SMTP session.
4800 54. The store-handling optimization for building very long strings was not
4801 differentiating between the different store pools. I don't think this
4802 actually made any difference in practice, but I've tidied it.
4804 55. While running the routers to verify a sender address, $sender_address
4805 was still set to the sender address. This is wrong, because when routing to
4806 send a bounce to the sender, it would be empty. Therefore, I have changed
4807 it so that, while verifying a sender address, $sender_address is set to <>.
4808 (There is no change to what happens when verifying a recipient address.)
4810 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
4811 target A or AAAA records (if not already returned) without resetting the
4812 qualify_single or search_parents options of the DNS resolver. These are
4813 inappropriate in this case because the targets of MX and SRV records must
4814 be FQDNs. A broken DNS record could cause trouble if it happened to have a
4815 target that, when qualified, matched something in the local domain. These
4816 two options are now turned off when doing these lookups.
4818 57. It seems that at least some releases of Reiserfs (which does not have the
4819 concept of a fixed number of inodes) returns zero and not -1 for the
4820 number of available inodes. This interacted badly with check_spool_inodes,
4821 which assumed that -1 was the "no such thing" setting. What I have done is
4822 to check that the total number of inodes is greater than zero before doing
4823 the test of how many are available.
4825 58. When a "warn" ACL statement has a log_message modifier, the message is
4826 remembered, and not repeated. This is to avoid a lot of repetition when a
4827 message has many recipients that cause the same warning to be written.
4828 Howewer, Exim was preserving the list of already written lines for an
4829 entire SMTP session, which doesn't seem right. The memory is now reset if a
4830 new message is started.
4832 59. The "rewrite" debugging flag was not showing the result of rewriting in the
4833 debugging output unless log_rewrite was also set.
4835 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
4836 of (int)(handle) when we know that handle contains (void *)(-1).
4838 61. The Exim daemon panic-logs an error return when it closes the incoming
4839 connection. However "connection reset by peer" seems to be common, and
4840 isn't really an error worthy of noting specially, so that particular error
4843 62. When Exim is trying to find all the local interfaces, it used to panic and
4844 die if the ioctl to get the interface flags failed. However, it seems that
4845 on at least one OS (Solaris 9) it is possible to have an interface that is
4846 included in the list of interfaces, but for which you get a failure error
4847 for this call. This happens when the interface is not "plumbed" into a
4848 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
4849 failure of the "get flags" call assumes that the interface is down.
4851 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
4852 makes life easier for people who are doing arithmetic on fields extracted
4853 from dates, where you often get leading zeros that should not be
4854 interpreted as octal.
4856 64. Added qualify_domain to the redirect router, to override the global
4859 65. If a pathologically long header line contained very many addresses (the
4860 report of this problem mentioned 10 000) and each of them was rewritten,
4861 Exim could use up a very large amount of memory. (It kept on making new
4862 copies of the header line as it rewrote, and never released the old ones.)
4863 At the expense of a bit more processing, the header rewriting function has
4864 been changed so that it no longer eats memory in this way.
4866 66. The generation of the Received: header has been moved from the time that a
4867 message starts to be received, to the time that it finishes. The timestamp
4868 in the Received: header should now be very close to that of the <= log
4869 line. There are two side-effects of this change:
4871 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
4872 logged header lines no longer include the local Received: line, because
4873 it has not yet been created. The same applies to a copy of the message
4874 that is returned to a non-SMTP sender when a message is rejected.
4876 (b) When a filter file is tested using -bf, no additional Received: header
4877 is added to the test message. After some thought, I decided that this
4880 This change does not affect the value of $received_for. It is still set
4881 after address rewriting, but before local_scan() is called.
4883 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
4885 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
4886 gave an unhelpful panic error message, and a defer error. I have managed to
4887 change this behaviour so that it now rejects any supplied certificate,
4888 which seems right, as the list of acceptable certificates is empty.
4890 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
4891 gave an unhelpful defer error. I have not managed to make this reject any
4892 supplied certificates, but the error message it gives is "no certificate
4893 supplied", which is not helpful.
4895 70. exigrep's output now also includes lines that are not associated with any
4896 message, but which match the given pattern. Implemented by a patch from
4897 Martin Sluka, which also tidied up the Perl a bit.
4899 71. Recipient callout verification, like sender verification, was using <> in
4900 the MAIL FROM command. This isn't really the right thing, since the actual
4901 sender may affect whether the remote host accepts the recipient or not. I
4902 have changed it to use the actual sender in the callout; this means that
4903 the cache record is now keyed on a recipient/sender pair, not just the
4904 recipient address. There doesn't seem to be a real danger of callout loops,
4905 since a callout by the remote host to check the sender would use <>.
4906 [SEE ABOVE: changed after hitting problems.]
4908 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
4909 temporary errors. However, in the case of such a code being given after
4910 the end of a data transmission (i.e. after ".") Exim was failing to write
4911 a retry record for the message. (Yes, there was some broken host that was
4912 actually sending 8xx at this point.)
4914 73. An unknown lookup type in a host list could cause Exim to panic-die when
4915 the list was checked. (An example that provoked this was putting <; in the
4916 middle of a list instead of at the start.) If this happened during a DATA
4917 ACL check, a -D file could be left lying around. This kind of configuration
4918 error no longer causes Exim to die; instead it causes a defer errror. The
4919 incident is still logged to the main and panic logs.
4921 74. Buglet left over from Exim 3 conversion. The message "too many messages
4922 in one connection" was written to the rejectlog but not the mainlog, except
4923 when address rewriting (yes!) was being logged.
4925 75. Added write_rejectlog option.
4927 76. When a system filter was run not as root (that is, when system_filter_user
4928 was set), the values of the $n variables were not being returned to the
4929 main process; thus, they were not subsequently available in the $sn
4932 77. Added +return_path_on_delivery log selector.
4934 78. A connection timeout was being treated differently from recipients deferred
4935 when testing hosts_max_try with a message that was older than the host's
4936 retry timeout. (The host should not be counted, thus allowing all hosts to
4937 be tried at least once before bouncing.) This may have been the cause of an
4938 occasionally reported bug whereby a message would remain on the queue
4939 longer than the retry timeout, but would be bounced if a delivery was
4940 forced. I say "may" because I never totally pinned down the problem;
4941 setting up timeout/retry tests is difficult. See also the next item.
4943 79. The ultimate address timeout was not being applied to errors that involved
4944 a combination of host plus message (for example, a timeout on a MAIL
4945 command). When an address resolved to a number of possible hosts, and they
4946 were not all tried for each delivery (e.g. because of hosts_max_try), a
4947 message could remain on the queue longer than the retry timeout.
4949 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4952 81. Fixed an obscure SMTP outgoing bug which required at least the following
4953 conditions: (a) there was another message waiting for the same server;
4954 (b) the server returned 5xx to all RCPT commands in the first message so
4955 that the message was not completed; (c) the server dropped the connection
4956 or gave a negative response to the RSET that Exim sends to abort the
4957 transaction. The observed case was a dropped connection after DATA that had
4958 been sent in pipelining mode. That is, the server had advertised PIPELINING
4959 but was not implementing it correctly. The effect of the bug was incorrect
4960 behaviour, such as trying another host, and this could lead to a crash.
4966 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4967 and daemon.c were passed as pointers to ints; they should have been
4968 pointers to socklen_t variables (which are typically unsigned ints).
4970 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4973 3. Fixed a really odd bug that affected only the testing scheme; patching a
4974 certain fixed string in the binary changed the value of another string that
4975 happened to be identical to the end of the original first string.
4977 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4978 name", it returns that address as the IP address. On some operating
4979 systems (e.g. Solaris), it also passes back the IP address string as the
4980 "host name". However, on others (e.g. Linux), it passes back an empty
4981 string. Exim wasn't checking for this, and was changing the host name to an
4982 empty string, assuming it had been canonicized.
4984 5. Although rare, it is permitted to have more than one PTR record for a given
4985 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4986 all the names associated with an address, because they do in Solaris.
4987 However, it seems that they do not in Linux for data that comes from the
4988 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4989 I found this out when I moved to a new Linux workstation and tried to run
4990 the Exim test suite.
4992 To get round this problem I have changed the code so that it now does its
4993 own call to the DNS to look up PTR records when searching for a host name.
4994 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
4995 addresses that are only in /etc/hosts are still found.
4997 This behaviour is, however, controlled by an option called host_lookup_
4998 order, which defaults to "bydns:byaddr". If people want to use the other
4999 order, or indeed, just use one or the other means of lookup, they can
5000 specify it in this variable.
5002 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5003 some operating systems, this comes back from gethostbyaddr() as an empty
5004 string, and this is what Exim used to test for. However, it seems that in
5005 other systems, "." is yielded. Exim now tests for this case too.
5007 7. The values of check_spool_space and check_log_space are now held internally
5008 as a number of kilobytes instead of an absolute number of bytes. If a
5009 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5010 kilobyte. This means that much larger values can be stored.
5012 8. Exim monitor: an attempt to get the action menu when not actually pointing
5013 at a message produces an empty menu entitled "No message selected". This
5014 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5015 no entries in it ("Shell widget menu has zero width and/or height"). So I
5016 have added a single, blank menu entry in this case.
5018 9. Added ${quote_local_part.
5020 10. MIME decoding is now applied to the contents of Subject: header lines when
5023 11. Now that a reference to $sender_host_address automatically causes a reverse
5024 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5025 host lookup before query-style lookups in lists that might use this
5026 variable. This has therefore been abolished, and the "net-" prefix is no
5027 longer necessary for query-style lookups.
5029 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5030 have been a typo for LFLAGS, so it has been changed.
5032 13. The install script calls Exim with "-C /dev/null" in order to find the
5033 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5034 to be output. Howeve, since Exim outputs its version number before the
5035 error, it didn't break the script. It just looked ugly. I fixed this by
5036 always allowing "-C /dev/null" if the caller is root.
5038 14. Ignore overlarge ACL variable number when reading spool file - insurance
5039 against a later release with more variables having written the file.
5041 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5042 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5043 this, as well as the form without the "IPv6" on the front.
5045 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5046 OS/Makefile-Darwin file.
5048 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5049 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5051 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5052 variable. However, some OS use ints for inodes. I've added cast to long int
5053 to get rid of the compiler warning.
5055 19. I had forgotten to lock out "/../" in configuration file names when
5056 ALT_CONFIG_PREFIX was set.
5058 20. Routers used for verification do not need to specify transports. However,
5059 if such a router generated a host list, and callout was configured, Exim
5060 crashed, because it could not find a port number from the (non-existent)
5061 transport. It now assumes port 25 in this circumstance.
5063 21. Added the -t option to exigrep.
5065 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5066 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5067 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5068 them is set, LOOKUP_LSEARCH is forced.
5070 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5072 24. Added sender and host information to the "rejected by local_scan()" log
5073 line; previously there was no indication of these.
5075 25. Added .include_if_exists.
5077 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5078 Linux. It turns out that not all file systems support this. Apparently some
5079 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5080 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5081 not supported on the file descriptor, is now ignored when Exim is trying to
5082 sync a directory. This applies only to Linux.
5084 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5086 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5087 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5090 29. Added support for maildirsize files from supplied patch (modified a bit).
5092 30. The use of :fail: followed by an empty string could lead Exim to respond to
5093 sender verification failures with (e.g.):
5095 550 Verification failed for <xxx>
5096 550 Sender verify failed
5098 where the first response line was missing the '-' that indicates it is not
5099 the final line of the response.
5101 31. The loop for finding the name of the user that called Exim had a hardwired
5102 limit of 10; it now uses the value of finduser_retries, which is used for
5103 all other user lookups.
5105 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5108 33. Exim was neglecting to zero errno before one call of strtol() when
5109 expanding a string and expecting an integer value. On some systems this
5110 resulted in spurious "integer overflow" errors. Also, it was casting the
5111 result into an int without checking.
5113 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5114 did not work. The code looks as if it has *never* worked, though it appears
5115 to have been documented since at least releast 1.62. I have made it work.
5117 35. The "timeout_DNS" error in retry rules, also documented since at least
5118 1.62, also never worked. As it isn't clear exactly what this means, and
5119 clearly it isn't a major issue, I have abolished the feature by treating it
5120 as "timeout", and writing a warning to the main and panic logs.
5122 36. The display of retry rules for -brt wasn't always showing the error code
5125 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5126 timeout_connect_A, timeout_connect_MX.
5128 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5129 to the empty sender.
5131 39. The daemon was not analysing the content of -oX till after it had closed
5132 stderr and disconnected from the controlling terminal. This meant that any
5133 syntax errors were only noted on the panic log, and the return code from
5134 the command was 0. By re-arranging the code a little, I've made the
5135 decoding happen first, so such errors now appear on stderr, and the return
5136 code is 1. However, the actual setting up of the sockets still happens in
5137 the disconnected process, so errors there are still only recorded on the
5140 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5141 connections (as happens on some IP stacks) was logged at start up time as
5142 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5143 it from "IPv6 and IPv4", which means that two separate sockets are being
5146 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5147 says whether AF_INET or AF_INET6 was passed as an argument.
5149 42. Exiwhat output was messed up when time zones were included in log
5152 43. Exiwhat now gives more information about the daemon's listening ports,
5153 and whether -tls-on-connect was used.
5155 44. The "port" option of the smtp transport is now expanded.
5157 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5158 silently ignored. Now an error message is written to the main and panic
5161 46. There's a new ACL modifier called "logwrite" which writes to a log file
5162 as soon as it is encountered.
5164 47. Added $local_user_uid and $local_user_gid at routing time.
5166 48. Exim crashed when trying to verify a sender address that was being
5169 49. Exim was recognizing only a space character after ".include". It now also
5170 recognizes a tab character.
5172 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5173 extracting the relevant information from the specification. The man page no
5174 longer contains scrambled data for the -d option, and I've added a section
5175 at the front about calling Exim under different names.
5177 51. Added "extra_headers" argument to the "mail" command in filter files.
5179 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5182 53. Installed eximstats 1.29.
5184 54. Added transport_filter_timeout as a generic transport option.
5186 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5187 Cc: header lines. This was required by RFC 822, but it not required by RFC
5190 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5191 incoming messages that did not have them. Now it does so only if the
5192 message originates locally, that is, if there is no associated remote host
5193 address. When Resent- header lines are present, this applies to the Resent-
5194 lines rather than the non-Resent- lines.
5196 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5197 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5199 58. Messages for configuration errors now include the name of the main
5200 configuration file - useful now that there may be more than one file in a
5201 list (.included file names were always shown).
5203 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5204 for those rare installations that do not start the daemon as root or run it
5205 setuid root. I've cut out the call to initgroups() if the daemon is not
5208 60. The Exim user and group can now be bound into the binary as text strings
5209 that are looked up at the start of Exim's processing.
5211 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5213 62. Added $mailstore_basename variable.
5215 63. Installed patch to sieve.c from Michael Haardt.
5217 64. When Exim failed to open the panic log after failing to open the main log,
5218 the original message it was trying to log was written to stderr and debug
5219 output, but if they were not available (the usual case in production), it
5220 was lost. Now it is written to syslog before the two lines that record the
5221 failures to open the logs.
5223 65. Users' Exim filters run in subprocesses under the user's uid. It is
5224 possible for a "deliver" command or an alias in a "personal" command to
5225 provoke an address rewrite. If logging of address rewriting is configured,
5226 this fails because the process is not running as root or exim. There may be
5227 a better way of dealing with this, but for the moment (because 4.30 needs
5228 to be released), I have disabled address rewrite logging when running a
5229 filter in a non-root, non-exim process.
5235 1. The buildconfig auxiliary program wasn't quoting the value set for
5236 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5237 not defined. This bug was masked in 4.22 by the effect that was fixed in
5240 2. Some messages that were rejected after a message id was allocated were
5241 shown as "incomplete" by exigrep. It no longer does this for messages that
5242 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5244 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5245 have allow_domain_literals set, the ID did not get logged in the <= line.
5246 Domain literals are now always recognized in Message-ID: header lines.
5248 4. The first argument for a ${extract expansion item is the key name or field
5249 number. Leading and trailing spaces in this item were not being ignored,
5250 causing some misleading effects.
5252 5. When deliver_drop_privilege was set, single queue runner processes started
5253 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5254 same command in the process it spins off) were not dropping privilege.
5256 6. When the daemon running as "exim" started a queue runner, it always
5257 re-executed Exim in the spun-off process. This is a waste of effort when
5258 deliver_drop_privilege is set. The new process now just calls the
5259 queue-runner function directly.
5265 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5268 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5269 ignored. Though the use of -oP was forcing the writing of a pid file, it
5270 was always written to the default place.
5272 3. If the message "no IP address found for host xxxx" is generated during
5273 incoming verification, it is now followed by identification of the incoming
5274 connection (so you can more easily find what provoked it).
5276 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5278 5. Added some features to "harden" Exim a bit more against certain attacks:
5280 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5281 be put in Local/Makefile. This is like the never_users runtime option,
5282 but it cannot be overridden. The default setting is "root".
5284 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5285 prefix string with which any file named in a -C command line option
5288 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5289 is retained for -C and -D only if the caller of Exim is root. Without
5290 it, the exim user may also use -C and -D and retain privilege.
5292 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5293 command line option is disabled.
5295 6. Macro names set by the -D option must start with an upper case letter, just
5296 like macro names defined in the configuration file.
5298 7. Added "dereference=" facility to LDAP.
5300 8. Two instances of the typo "uknown" in the source files are fixed.
5302 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5303 the Configure-Makefile script screwed up while processing it.
5305 10. Incorporated PCRE 4.4.
5307 11. The SMTP synchronization check was not operating right at the start of an
5308 SMTP session. For example, it could not catch a HELO sent before the client
5309 waited for the greeting. There is now a check for outstanding input at the
5310 point when the greeting is written. Because of the duplex, asynchronous
5311 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5312 way, but not yet received, when the check is performed.
5314 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5315 on TCP/IP sockets, because this apparently causes some broken clients to
5318 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5319 unchanged) from the Cygwin maintainer.
5321 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5322 was supported instead of testing for mbx. Effectively a typo.
5324 15. The spa authenticator server code was not checking that the input it
5325 received was valid base64.
5327 16. The debug output line for the "set" modifier in ACLs was not showing the
5328 name of the variable that was being set.
5330 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5332 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5333 lookup on its own. Something else was needed to trigger the lookup. For
5334 example, a match in host_lookup or the need for a host name in a host list.
5335 Now, if $sender_host_name is referenced and the host name has not yet been
5336 looked up, a lookup is performed. If the lookup fails, the variable remains
5337 empty, and $host_lookup_failed is set to "1".
5339 19. Added "eqi" as a case-independent comparison operator.
5341 20. The saslauthd authentication condition could segfault if neither service
5342 nor realm was specified.
5344 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5345 error message that was logged was misleading, and incoming SMTP
5346 connections were dropped. The message is now more accurate, and temporary
5347 errors are given to SMTP connections.
5349 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5350 (except RSET) with a 421 error, until QUIT is received. However, it was
5351 failing to send a response to QUIT.
5353 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5354 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5355 ACL is now run after the helo_try_verify_hosts code.
5357 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5358 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5359 case-independent, so other case variants are also recognized). Apparently
5360 some systems use these upper case variants.
5362 25. If more than two messages were waiting for the same host, and a transport
5363 filter was specified for the transport, Exim sent two messages over the
5364 same TCP/IP connection, and then failed with "socket operation on non-
5365 socket" when it tried to send the third.
5367 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5369 27. The extern definition of crypt16() in expand.c was not being excluded when
5370 the OS had its own crypt16() function.
5372 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5373 as a preferred synonym for return_size_limit, both as an option and as an
5376 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5378 30. Changed the default configuration ACL to relax the local part checking rule
5379 for addresses that are not in any local domains. For these addresses,
5380 slashes and pipe symbols are allowed within local parts, but the sequence
5381 /../ is explicitly forbidden.
5383 31. SPA server authentication was not clearing the challenge buffer before
5386 32. log_message in a "warn" ACL statement was writing to the reject log as
5387 well as to the main log, which contradicts the documentation and doesn't
5388 seem right (because no rejection is happening). So I have stopped it.
5390 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5391 However, I am unable to do any testing of this.
5393 34. Fixed an infelicity in the appendfile transport. When checking directories
5394 for a mailbox, to see if any needed to be created, it was accidentally
5395 using path names with one or more superfluous leading slashes; tracing
5396 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5398 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5399 contents are added to the log line that is written for every discarded
5400 recipient. (Previously a log_message setting was ignored.)
5402 36. The ${quote: operator now quotes the string if it is empty.
5404 37. The install script runs exim in order to find its version number. If for
5405 some reason other than non-existence or emptiness, which it checks, it
5406 could not run './exim', it was installing it with an empty version number,
5407 i.e. as "exim-". This error state is now caught, and the installation is
5410 38. An argument was missing from the function that creates an error message
5411 when Exim fails to connect to the socket for saslauthd authentication.
5412 This could cause Exim to crash, or give a corrupted message.
5414 39. Added isip, isip4, and isip6 to ${if conditions.
5416 40. The ACL variables $acl_xx are now saved with the message, and can be
5417 accessed later in routers, transports, and filters.
5419 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
5420 strings in the file are not string-expanded.
5422 42. If a MAIL command specified a SIZE value that was too large to fit into an
5423 int variable, the check against message_size_limit failed. Such values are
5424 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
5425 day this will have to be increased, but I don't think I want to be around
5426 when emails are that large.
5433 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
5434 iconv() is not standard in FreeBSD.
5436 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
5437 IPv6 enabled. The observed symptom was a segmentation fault on return from
5438 the function os_common_find_running_interfaces() in src/os.c.
5440 3. In the check_special_case() function in daemon.c I had used "errno" as an
5441 argument name, which causes warnings on some systems. This was basically a
5442 typo, since it was named "eno" in the comments!
5444 4. The code that waits for the clock to tick (at a resolution of some fraction
5445 of a second) so as to ensure message-id uniqueness was always waiting for
5446 at least one whole tick, when it could have waited for less. [This is
5447 almost certainly not relevant at current processor speeds, where it is
5448 unlikely to ever wait at all. But we try to future-proof.]
5450 5. The function that sleeps for a time interval that includes fractions of a
5451 second contained a race. It did not block SIGALRM between setting the
5452 timer, and suspending (a couple of lines later). If the interval was short
5453 and the sigsuspend() was delayed until after it had expired, the suspension
5454 never ended. On busy systems this could lead to processes getting stuck for
5457 6. Some uncommon configurations may cause a lookup to happen in a queue runner
5458 process, before it forks any delivery processes. The open lookup caching
5459 mechanism meant that the open file or database connection was passed into
5460 the delivery process. The problem was that delivery processes always tidy
5461 up cached lookup data. This could cause a problem for the next delivery
5462 process started by the queue runner, because the external queue runner
5463 process does not know about the closure. So the next delivery process
5464 still has data in the lookup cache. In the case of a file lookup, there was
5465 no problem because closing a file descriptor in a subprocess doesn't affect
5466 the parent. However, if the lookup was caching a connection to a database,
5467 the connection was closed, and the second delivery process was likely to
5468 see errors such as "PGSQL: query failed: server closed the connection
5469 unexpectedly". The problem has been fixed by closing all cached lookups
5470 in a queue runner before running a delivery process.
5472 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
5473 seem to have the "const" qualifier which it has on other OS. I've
5476 8. Change 4.21/2 was too strict. It is only if there are two authenticators
5477 *of the same type* (client or server) with the same public name that an
5478 error should be diagnosed.
5480 9. When Exim looked up a host name for an IP address, but failed to find the
5481 original IP address when looking up the host name (a safety check), it
5482 output the message "<ip address> does not match any IP for NULL", which was
5483 confusing, to say the least. The bug was that the host name should have
5484 appeared instead of "NULL".
5486 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
5487 user that is built into the binary, and the -C or -D options is used, root
5488 privilege is dropped before the configuration file is read. In addition,
5489 logging is switched to stderr instead of the normal log files. If the
5490 configuration then re-defines the Exim user, the unprivileged environment
5491 is probably not what is expected, so Exim logs a panic warning message (but
5494 However, if deliver_drop_privilege is set, the unprivileged state may well
5495 be exactly what is intended, so the warning has been cut out in that case,
5496 and Exim is allowed to try to write to its normal log files.
5502 1. smtp_return_error_details was not giving details for temporary sender
5503 or receiver verification errors.
5505 2. Diagnose a configuration error if two authenticators have the same public
5508 3. Exim used not to create the message log file for a message until the first
5509 delivery attempt. This could be confusing when incoming messages were held
5510 for policy or load reasons. The message log file is now created at the time
5511 the message is received, and an initial "Received" line is written to it.
5513 4. The automatically generated man page for command line options had a minor
5514 bug that caused no ill effects; however, a more serious problem was that
5515 the procedure for building the man page automatically didn't always
5516 operate. Consequently, release 4.20 contains an out-of-date version. This
5517 shouldn't happen again.
5519 5. When building Exim with embedded Perl support, the script that builds the
5520 Makefile was calling 'perl' to find its compile-time parameters, ignoring
5521 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
5523 6. The freeze_tell option was not being used for messages that were frozen on
5524 arrival, either by an ACL or by local_scan().
5526 7. Added the smtp_incomplete_transaction log selector.
5528 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
5529 was accepting AUTH without a new EHLO.
5531 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
5532 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
5535 10. Logging of TCP/IP connections (when configured) now happens in the main
5536 daemon process instead of the child process, so that the TCP/IP connection
5537 count is more accurate (but it can never be perfect).
5539 11. The use of "drop" in a nested ACL was not being handled correctly in the
5540 outer ACL. Now, if condition failure induced by the nested "drop" causes
5541 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
5542 or "require"), the connection is dropped.
5544 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
5545 that yield "discard" can now be used with an "accept" or a "discard" verb,
5546 but an error is generated for any others (because I can't see a useful way
5547 to define what should happen).
5549 13. When an ACL is read dynamically from a file (or anywhere else), the lines
5550 are now processed in the same way as lines in the Exim configuration file.
5551 In particular, continuation lines are supported.
5553 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
5555 15. Added -ti meaning -t -i.
5557 16. Check for letters, digits, hyphens, and dots in the names of dnslist
5558 domains, and warn by logging if others are found.
5560 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
5561 returned from GIFCONF when Exim is trying to find the list of interfaces on
5562 a host. The code in os.c has been modified to copy each ifreq to an aligned
5563 structure in all cases.
5565 Also, in some cases, the returned ifreq's were being copied to a 'struct
5566 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
5567 means the last couple of bytes of an IPv6 address could be chopped if the
5568 ifreq contained only a normal sockaddr (14 bytes storage).
5570 18. Named domain lists were not supported in the hosts_treat_as_local option.
5571 An entry such as +xxxx was not recognized, and was treated as a literal
5574 19. Ensure that header lines added by a DATA ACL are included in the reject log
5575 if the ACL subsequently rejects the message.
5577 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
5578 MD5 (which is deprecated).
5580 21. When testing a filter file using -bf, Exim was writing a message when it
5581 took the sender from a "From " line in the message, but it was not doing so
5582 when it took $return_path from a Return-Path: header line. It now does.
5584 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
5585 with a valid header line field name (a series of printing characters
5586 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
5588 23. Changed "disc" in the source to "disk" to conform to the documentation and
5589 the book and for uniformity.
5591 24. Ignore Sendmail's -Ooption=value command line item.
5593 25. When execve() failed while trying to run a command in a pipe transport,
5594 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
5595 could be confused with a return value of 69 from the command itself. This
5596 has been changed to 127, the value the shell returns if it is asked to run
5597 a non-existent command. The wording for the related log line suggests a
5598 non-existent command as the problem.
5600 26. If received_header_text expands to an empty string, do not add a Received:
5601 header line to the message. (Well, it adds a token one on the spool, but
5602 marks it "old" so that it doesn't get used or transmitted.)
5604 27. Installed eximstats 1.28 (addition of -nt option).
5606 28. There was no check for failure on the call to getsockname() in the daemon
5607 code. This can fail if there is a shortage of resources on the system, with
5608 ENOMEM, for example. A temporary error is now given on failure.
5610 29. Contrary to the C standard, it seems that in some environments, the
5611 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
5612 program. Exim now does this explicitly; it affects the formatting of
5613 timestamps using strftime().
5615 30. If exiqsumm was given junk data, it threw up some uninitialized variable
5616 complaints. I've now initialized all the variables, to avoid this.
5618 32. Header lines added by a system filter were not being "seen" during
5619 transport-time rewrites.
5621 33. The info_callback() function passed to OpenSSL is set up with type void
5622 (*)(SSL *, int, int), as described somewhere. However, when calling the
5623 function (actually a macro) that sets it up, the type void(*)() is
5624 expected. I've put in a cast to prevent warnings from picky compilers.
5626 34. If a DNS black list lookup found a CNAME record, but there were no A
5627 records associated with the domain it pointed at, Exim crashed.
5629 35. If a DNS black list lookup returned more than one A record, Exim ignored
5630 all but the first. It now scans all returned addresses if a particular IP
5631 value is being sought. In this situation, the contents of the
5632 $dnslist_value variable are a list of all the addresses, separated by a
5635 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
5636 to accept a host name and all its aliases if the forward lookup for any of
5637 them yielded the IP address of the incoming connection. Now it accepts only
5638 those names whose forward lookup yields the correct IP address. Any other
5639 names are discarded. This closes a loophole whereby a rogue DNS
5640 administrator could create reverse DNS records to break through a
5641 wildcarded host restriction in an ACL.
5643 37. If a user filter or a system filter that ran in a subprocess used any of
5644 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
5645 the wrong values were passed to the pipe command ($thisaddress had the
5646 value of $0, $0 had the value of $1, etc). This bug was introduced by
5647 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
5649 38. Improved the line breaking for long SMTP error messages from ACLs.
5650 Previously, if there was no break point between 40 and 75 characters, Exim
5651 left the rest of the message alone. Two changes have been made: (a) I've
5652 reduced the minimum length to 35 characters; (b) if it can't find a break
5653 point between 35 and 75 characters, it looks ahead and uses the first one
5654 that it finds. This may give the occasional overlong line, but at least the
5655 remaining text gets split now.
5657 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
5658 file descriptors might be low, and that setting 1000 would always raise it.
5659 It turns out that in some environments, the limit is already over 1000 and
5660 that lowering it causes trouble. So now Exim takes care not to decrease it.
5662 40. When delivering a message, the value of $return_path is set to $sender_
5663 address at the start of routing (routers may change the value). By an
5664 oversight, this default was not being set up when an address was tested by
5665 -bt or -bv, which affected the outcome if any router or filter referred to
5668 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
5669 log only when "message" or "log_message" are set. However, if one of the
5670 conditions was an address verification, or a call to a nested ACL, the
5671 messages generated by the underlying test were being passed through. This
5672 no longer happens. The underlying message is available in $acl_verify_
5673 message for both "message" and "log_message" expansions, so it can be
5674 passed through if needed.
5676 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
5677 new expansion $bh_ to give the encoded byte string without charset
5678 translation. Translation happens only if iconv() is available; HAVE_ICONV
5679 indicates this at build time. HEADERS_CHARSET gives the charset to
5680 translate to; headers_charset can change it in the configuration, and
5681 "headers charset" can change it in an individual filter file.
5683 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
5684 that creates RFC 2047 encoded "words" labels them as that charset instead
5685 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
5686 expansion operator; (ii) when Exim creates a From: line for a local
5687 message; (iii) when a header line is rewritten to include a "phrase" part.
5689 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
5690 buggy, causing it to skip the first lines of messages whose message ID
5691 ended in 'D'. This would not have bitten before Exim release 4.14, because
5692 message IDs were unlikely to end in 'D' before then. The effect was to have
5693 incorrect size information for certain domains.
5695 45. #include "config.h" was missing at the start of the crypt16.c module. This
5696 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
5699 46. If there was a timeout during a "random" callout check, Exim treated it as
5700 a failure of the random address, and carried on sending RSET and the real
5701 address. If the delay was just some slowness somewhere, the response to the
5702 original RCPT would be taken as a response to RSET and so on, causing
5703 mayhem of various kinds.
5705 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
5706 when I implemented it. It didn't allow for the fact that some option values
5707 may legitimatetly be negative (e.g. size_addition), and it didn't even do
5708 the right test for positive values.
5710 48. Domain names in DNS records are case-independent. Exim always looks them up
5711 in lower case. Some resolvers return domain names in exactly the case they
5712 appear in the zone file, that is, they may contain uppercase letters. Not
5713 all resolvers do this - some return always lower case. Exim was treating a
5714 change of case by a resolver as a change of domain, similar to a widening
5715 of a domain abbreviation. This triggered its re-routing code and so it was
5716 trying to route what was effectively the same domain again. This normally
5717 caused routing to fail (because the router wouldn't handle the domain
5718 twice). Now Exim checks for this case specially, and just changes the
5719 casing of the domain that it ultimately uses when it transmits the message
5722 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
5725 50. If a filter generated a file delivery with a non-absolute name (possible if
5726 no home directory exists for the router), the forbid_file option was not
5729 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
5730 the existing equality matching.
5732 52. Exim was using ints instead of ino_t variables in some places where it was
5733 dealing with inode numbers.
5735 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
5736 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
5737 called TMPDIR, and if it finds it is different, it changes its value.
5739 54. The smtp_printf() function is now made available to local_scan() so
5740 additional output lines can be written before returning. There is also an
5741 smtp_fflush() function to enable the detection of a dropped connection.
5742 The variables smtp_input and smtp_batched_input are exported to
5745 55. Changed the default runtime configuration: the message "Unknown user"
5746 has been removed from the ACL, and instead placed on the localuser router,
5747 using the cannot_route_message feature. This means that any verification
5748 failures that generate their own messages won't get overridden. Similarly,
5749 the "Unrouteable address" message that was in the ACL for unverifiable
5750 relay addresses has also been removed.
5752 56. Added hosts_avoid_esmtp to the smtp transport.
5754 57. The exicyclog script was not checking for the esoteric option
5755 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
5756 will work only if exicyclog is run under the appropriate euid.
5758 58. Following a discussion on the list, the rules by which Exim recognises line
5759 endings on incoming messages have been changed. The -dropcr and drop_cr
5760 options are now no-ops, retained only for backwards compatibility. The
5761 following line terminators are recognized: LF CRLF CR. However, special
5762 processing applies to CR:
5764 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
5765 nor a local message in the state where . is a terminator.
5767 (ii) If a bare CR is encountered in a header line, an extra space is added
5768 after the line terminator so as not to end the header. The reasoning
5769 behind this is that bare CRs in header lines are most likely either
5770 to be mistakes, or people trying to play silly games.
5772 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
5773 was being incorrectly given as 18 bytes larger than it should have been.
5774 This is a VOB (very old bug).
5776 60. This may never have affected anything current, but just in case it has:
5777 When the local host is found other than at the start of a list of hosts,
5778 the local host, those with the same MX, and any that follow, are discarded.
5779 When the list in question was part of a longer list of hosts, the following
5780 hosts (not currently being processed) were also being discarded. This no
5781 longer happens. I'm not sure if this situation could ever has previously
5784 61. Added the "/MX" feature to lists of hosts in the manualroute and query
5787 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
5788 header. This is something that is recommended in a new Internet Draft, and
5789 is something that is documented as being done by Sendmail. There are two
5790 possible values. For messages generated by the autoreply transport, Exim
5793 Auto-Submitted: auto-replied
5795 whereas for all other generated messages (e.g. bounces) it adds
5797 Auto-Submitted: auto-generated
5799 63. The "personal" condition in filters now includes a test for the
5800 Auto-Submitted: header. If it contains the string "auto-" the message it
5801 not considered personal.
5803 64. Added rcpt_include_affixes as a generic transport option.
5805 65. Added queue_only_override (default true).
5807 66. Added the syslog_duplication option.
5809 67. If what should have been the first header line of a message consisted of
5810 a space followed by a colon, Exim was mis-interpreting it as a header line.
5811 It isn't of course - it is syntactically invalid and should therefore be
5812 treated as the start of the message body. The misbehaviour could have
5813 caused a number of strange effects, including loss of data in subsequent
5814 header lines, and spool format errors.
5816 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
5817 client host had authenticated. This control can now be exercised by an ACL
5818 for more flexibility.
5820 69. By default, callouts do not happen when testing with -bh. There is now a
5821 variant, -bhc, which does actually run the callout code, including
5822 consulting and updating the callout cache.
5824 70. Added support for saslauthd authentication, courtesy of Alexander
5827 71. If statvfs() failed on the spool or log directories while checking their
5828 size for availability, Exim confusingly gave the error "space shortage".
5829 Furthermore, in debugging mode it crashed with a floating point exception.
5830 These checks are done if check_{spool,log}_{space,inodes} are set, and when
5831 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
5832 serious problem, Exim now writes to the main and panic logs when this
5833 happens, with details of the failure. It then refuses to accept the
5834 incoming message, giving the message "spool directory problem" or "log
5835 directory problem" with a 421 code for SMTP messages.
5837 72. When Exim is about to re-exec itself, it ensures that the file descriptors
5838 0, 1, and 2 exist, because some OS complain for execs without them (see
5839 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
5840 descriptors. However, the code omitted to check that the open succeeded,
5841 causing mysterious errors if for some reason the permissions on /dev/null
5842 got screwed. Now Exim writes a message to the main and panic logs, and
5843 bombs out if it can't open /dev/null.
5845 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
5846 interact so that it is all more flexible. It is supposed to remain
5847 backwards compatible. Also added extra_local_interfaces.
5849 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
5850 to bomb out with an assertion failure - to the client this appears as a
5851 connection drop. This problem occurs in the part of the code that was taken
5852 from the Samba project. Fortunately, the assertion is in a very simple
5853 function, so I have fixed this by reproducing the function inline in the
5854 one place where it is called, and arranging for authentication to fail
5855 instead of killing the process with assert().
5857 75. The SPA client code was not working when the server requested OEM rather
5858 than Unicode encoding.
5860 76. Added code to make require_files with a specific uid setting more usable in
5861 the case where statting the file as root fails - usually a non-root-mounted
5862 NFS file system. When this happens and the failure is EACCES, Exim now
5863 forks a subprocess and does the per-uid checking as the relevant uid.
5865 77. Added process_log_path.
5867 78. If log_file_path was not explicitly set, a setting of check_log_space or
5868 check_log_inodes was ignored.
5870 79. If a space check for the spool or log partitions fails, the incident is now
5871 logged. Of course, in the latter case the data may get lost...
5873 80. Added the %p formatting code to string_format() so that it can be used to
5874 print addresses in debug_print(). Adjusted all the address printing in the
5875 debugging in store.c to use %p rather than %d.
5877 81. There was a concern that a line of code in smtp_in.c could overflow a
5878 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
5879 initially expressed, the concern was not well-founded, because trailing
5880 spaces are removed early. However, if the trailing spaces were followed by
5881 a NULL, they did not get removed, so the overflow was possible. Two fixes
5884 (a) I re-wrote the offending code in a cleaner fashion.
5885 (b) If an incoming SMTP command contains a NULL character, it is rejected
5888 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
5889 runs initgroups(), so that if the Exim user is in any additional groups,
5890 they will be used during message reception.
5896 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob