1 Change log file for Exim from version 4.21
2 -------------------------------------------
8 TF/01 Correctly close the server side of TLS when forking for delivery.
10 When a message was received over SMTP with TLS, Exim failed to clear up
11 the incoming connection properly after forking off the child process to
12 deliver the message. In some situations the subsequent outgoing
13 delivery connection happened to have the same fd number as the incoming
14 connection previously had. Exim would try to use TLS and fail, logging
15 a "Bad file descriptor" error.
17 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
18 utilities have not been installed.
20 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
21 temporary space as the ACL may create new global variables.
23 TL/01 LDAP support uses per connection or global context settings, depending
24 upon the detected version of the libraries at build time.
26 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
27 to extract and use the src ip:port in logging and expansions as if it
28 were a direct connection from the outside internet. PPv2 support was
29 updated based on HAProxy spec change in May 2014.
31 JH/02 Add ${listextract {number}{list}{success}{fail}}.
33 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
34 Properly escape header and check for NULL return.
36 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
39 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
41 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
42 characters in header names, implemented as a verify condition.
43 Contributed by Michael Fischer v. Mollard.
45 TL/05 Rename SPF condition results err_perm and err_temp to standardized
46 results permerror and temperror. Previous values are deprecated but
47 still accepted. In a future release, err_perm and err_temp will be
48 completely removed, which will be a backward incompatibility if the
49 ACL tests for either of these two old results. Patch contributed by
50 user bes-internal on the mailing list.
52 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
54 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
55 selectors, in both main and reject logs.
57 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
60 JH/07 Add malware type "sock" for talking to simple daemon.
62 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
64 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
65 routers/transports under cutthrough routing.
67 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
68 numbers. Touch up "bool" conditional to keep the same definition.
70 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
72 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
74 JH/12 Expand items in router/transport headers_add or headers_remove lists
75 individually rather than the list as a whole. Bug 1452.
77 Required for reasonable handling of multiple headers_ options when
78 they may be empty; requires that headers_remove items with embedded
79 colons must have them doubled (or the list-separator changed).
81 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
82 view the policy declared in the DMARC record. Currently, $dmarc_status
83 is a combined value of both the record presence and the result of the
86 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
88 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
89 dnslookup router and the smtp transport (applying to the forward
92 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
93 of ldap servers used for a specific lookup. Patch provided by Heiko
96 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
97 New variable $lookup_dnssec_authenticated for observability.
99 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
100 Patch submitted by Lars Timman.
102 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
104 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
105 Requires trusted mode and valid format message id, aborts otherwise.
106 Patch contributed by Heiko Schlichting.
108 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
109 certextract with support for various fields. Bug 1358.
111 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
112 is requested by default, modifiable by smtp transport option
115 JH/22 Expansion operators ${md5:string} and ${sha1::string} can now
116 operate on certificate variables to give certificate fingerprints
117 Also new ${sha256:cert_variable}.
119 JH/23 The PRDR feature is moved from being Experimental into the mainline.
121 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
122 Christian Aistleitner.
128 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
130 PP/02 Make -n do something, by making it not do something.
131 When combined with -bP, the name of an option is not output.
133 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
136 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
137 $sender_host_name and config options to manage this, and basic check
140 PP/05 DSCP support for outbound connections and control modifier for inbound.
142 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
143 (Only plugin which currently uses this is kerberos4, which nobody should
144 be using, but we should make it available and other future plugins might
145 conceivably use it, even though it would break NAT; stuff *should* be
146 using channel bindings instead).
148 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
149 name; added for Sendmail compatibility; requires admin caller.
150 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
151 just ignore it); requires trusted caller.
152 Also parse but ignore: -Ac -Am -X<logfile>
155 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
157 TL/02 Add +smtp_confirmation as a default logging option.
159 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
160 Patch by Magnus Holmgren from 2007-02-20.
162 TL/04 Bugzilla 1281 - Spec typo.
163 Bugzilla 1283 - Spec typo.
164 Bugzilla 1290 - Spec grammar fixes.
166 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
168 TL/06 Add Experimental DMARC support using libopendmarc libraries.
170 TL/07 Fix an out of order global option causing a segfault. Reported to dev
171 mailing list by by Dmitry Isaikin.
173 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
175 JH/02 Support "G" suffix to numbers in ${if comparisons.
177 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
179 NM/01 Bugzilla 1197 - Spec typo
180 Bugzilla 1196 - Spec examples corrections
182 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
184 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
185 gnutls_enable_pkcs11, but renamed to more accurately indicate its
188 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
189 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
191 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
192 "acl {{name}{arg}...}", and optional args on acl condition
195 JH/05 Permit multiple router/transport headers_add/remove lines.
197 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
199 JH/07 Avoid using a waiting database for a single-message-only transport.
200 Performance patch from Paul Fisher. Bugzilla 1262.
202 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
205 JH/09 Add $headers_added variable, with content from use of ACL modifier
206 add_header (but not yet added to the message). Bugzilla 199.
208 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
209 Pulled from Bugzilla 817 by Wolfgang Breyha.
211 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
213 (nb: this is the same fix as in Exim 4.80.1)
215 JH/11 Add A= logging on delivery lines, and a client_set_id option on
218 JH/12 Add optional authenticated_sender logging to A= and a log_selector
221 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
223 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
224 advertise SMTP AUTH mechanism to us, instead of a generic
225 protocol violation error. Also, make Exim more robust to bad
226 data from the Dovecot auth socket.
228 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
230 When a queue runner is handling a message, Exim first routes the
231 recipient addresses, during which it prunes them based on the retry
232 hints database. After that it attempts to deliver the message to
233 any remaining recipients. It then updates the hints database using
236 So if a recipient address works intermittently, it can get repeatedly
237 deferred at routing time. The retry hints record remains fresh so the
238 address never reaches the final cutoff time.
240 This is a fairly common occurrence when a user is bumping up against
241 their storage quota. Exim had some logic in its local delivery code
242 to deal with this. However it did not apply to per-recipient defers
243 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
245 This change adds a proper retry rule check during routing so that the
246 final cutoff time is checked against the message's age. We only do
247 this check if there is an address retry record and there is not a
248 domain retry record; this implies that previous attempts to handle
249 the address had the retry_use_local_parts option turned on. We use
250 this as an approximation for the destination being like a local
251 delivery, as in LMTP.
253 I suspect this new check makes the old local delivery cutoff check
254 redundant, but I have not verified this so I left the code in place.
256 TF/02 Correct gecos expansion when From: is a prefix of the username.
258 Test 0254 submits a message to Exim with the header
262 When I ran the test suite under the user fanf2, Exim expanded
263 the header to contain my full name, whereas it should have added
264 a Resent-Sender: header. It erroneously treats any prefix of the
265 username as equal to the username.
267 This change corrects that bug.
269 GF/01 DCC debug and logging tidyup
270 Error conditions log to paniclog rather than rejectlog.
271 Debug lines prefixed by "DCC: " to remove any ambiguity.
273 TF/03 Avoid unnecessary rebuilds of lookup-related code.
275 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
276 Bug spotted by Jeremy Harris; was flawed since initial commit.
277 Would have resulted in OCSP responses post-SNI triggering an Exim
278 NULL dereference and crash.
280 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
282 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
283 Bug detection, analysis and fix by Samuel Thibault.
284 Bugzilla 1331, Debian bug #698092.
286 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
288 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
289 Server implementation by Todd Lyons, client by JH.
290 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
291 config variable "prdr_enable" controls whether the server
292 advertises the facility. If the client requests PRDR a new
293 acl_data_smtp_prdr ACL is called once for each recipient, after
294 the body content is received and before the acl_smtp_data ACL.
295 The client is controlled by bolth of: a hosts_try_prdr option
296 on the smtp transport, and the server advertisement.
297 Default client logging of deliveries and rejections involving
298 PRDR are flagged with the string "PRDR".
300 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
301 fclose(). Diagnosis by Todd Lyons.
303 PP/17 Update configure.default to handle IPv6 localhost better.
304 Patch by Alain Williams (plus minor tweaks).
307 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
308 This is now consistent with GnuTLS, and is now documented: the
309 previous undocumented portable approach to treating the option as
310 unset was to force an expansion failure. That still works, and
311 an empty string is now equivalent.
313 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
314 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
315 not performing validation itself.
317 PP/20 Added force_command boolean option to pipe transport.
318 Patch from Nick Koston, of cPanel Inc.
320 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
323 TF/04 Added udpsend ACL modifer and hexquote expansion operator
325 PP/21 Fix eximon continuous updating with timestamped log-files.
326 Broken in a format-string cleanup in 4.80, missed when I repaired the
327 other false fix of the same issue.
328 Report and fix from Heiko Schlichting.
331 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
332 Report from Prashanth Katuri.
334 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
335 It's SecureTransport, so affects any MacOS clients which use the
336 system-integrated TLS libraries, including email clients.
338 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
339 using a MIME ACL for non-SMTP local injection.
340 Report and assistance in diagnosis by Warren Baker.
342 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
344 JH/16 Fix comparisons for 64b. Bugzilla 1385.
346 TL/09 Add expansion variable $authenticated_fail_id to keep track of
347 last id that failed so it may be referenced in subsequent ACL's.
349 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
352 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
353 ldap library initialization, allowing self-signed CA's to be
354 used. Also properly sets require_cert option later in code by
355 using NULL (global ldap config) instead of ldap handle (per
356 session). Bug diagnosis and testing by alxgomz.
358 TL/12 Enhanced documentation in the ratelimit.pl script provided in
359 the src/util/ subdirectory.
361 TL/13 Bug 1301 - Imported transport SQL logging patch from Axel Rau
362 renamed to Transport Post Delivery Action by Jeremy Harris, as
365 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
366 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
367 redis_servers = needs to be configured which will be used by the redis
368 lookup. Patch from Warren Baker, of The Packet Hub.
370 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
372 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
373 hostname or reverse DNS when processing a host list. Used suggestions
374 from multiple comments on this bug.
376 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
378 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
379 Missed a few lines, added it to make the runtest require no keyboard
382 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
383 contains upper case chars. Make router use caseful_local_part.
385 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
386 support when GnuTLS has been built with p11-kit.
392 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
394 This, or similar/improved, will also be change PP/11 of 4.82.
400 PP/01 Handle short writes when writing local log-files.
401 In practice, only affects FreeBSD (8 onwards).
402 Bugzilla 1053, with thanks to Dmitry Isaikin.
404 NM/01 Bugzilla 949 - Documentation tweak
406 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
409 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
411 PP/02 Implemented gsasl authenticator.
413 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
415 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
416 `pkg-config foo` for cflags/libs.
418 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
419 with rest of GSASL and with heimdal_gssapi.
421 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
422 `pkg-config foo` for cflags/libs for the TLS implementation.
424 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
425 properties get this fed in as external SSF. A number of robustness
426 and debugging improvements to the cyrus_sasl authenticator.
428 PP/08 cyrus_sasl server now expands the server_realm option.
430 PP/09 Bugzilla 1214 - Log authentication information in reject log.
431 Patch by Jeremy Harris.
433 PP/10 Added dbmjz lookup type.
435 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
437 PP/12 MAIL args handles TAB as well as SP, for better interop with
438 non-compliant senders.
439 Analysis and variant patch by Todd Lyons.
441 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
442 Bug report from Lars Müller <lars@samba.org> (via SUSE),
443 Patch from Dirk Mueller <dmueller@suse.com>
445 PP/13 tls_peerdn now print-escaped for spool files.
446 Observed some $tls_peerdn in wild which contained \n, which resulted
447 in spool file corruption.
449 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
450 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
451 or write after TLS renegotiation, which otherwise led to messages
454 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
455 as a tracking header (ie: a signed header comes before the signature).
456 Patch from Wolfgang Breyha.
458 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
459 comma-sep list; embedded commas doubled.
461 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
463 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
465 Report and patch from Dmitry Banschikov.
467 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
468 Removed SSL_clear() after SSL_new() which led to protocol negotiation
469 failures. We appear to now support TLS1.1+ with Exim.
471 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
472 lets Exim select keys and certificates based upon TLS SNI from client.
473 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
474 before an outbound SMTP session. New log_selector, +tls_sni.
476 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
477 NULL dereference. Report and patch from Alun Jones.
479 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
480 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
481 Not seeing resolver debug output on NetBSD, but suspect this is a
482 resolver implementation change.
484 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
485 Left warnings. Added "eximon gdb" invocation mode.
487 PP/21 Defaulting "accept_8bitmime" to true, not false.
489 PP/22 Added -bw for inetd wait mode support.
491 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
492 locate the relevant includes and libraries. Made this the default.
494 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
495 Bugzilla 1246, report and most of solution from Tomasz Kusy.
497 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
498 This may cause build issues on older platforms.
500 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
501 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
502 gnutls_require_mac & gnutls_require_protocols (no longer supported).
503 Added SNI support via GnuTLS too.
504 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
506 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
508 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
509 Applied second patch from Janne, implementing suggestion to default
510 multiple-strings-in-record handling to match SPF spec.
512 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
514 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
515 read-only, out of scope).
516 Patch from Wolfgang Breyha, report from Stuart Northfield.
518 PP/29 Fix three issues highlighted by clang analyser static analysis.
519 Only crash-plausible issue would require the Cambridge-specific
520 iplookup router and a misconfiguration.
521 Report from Marcin Mirosław.
523 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
525 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
526 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
527 As part of this, removing so much warning spew let me fix some minor
528 real issues in debug logging.
530 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
531 assignment on my part. Fixed.
533 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
534 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
535 Janne Snabb (who went above and beyond: thank you).
537 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
538 string otherwise requires a connection and a bunch more work and it's
539 relatively easy to get wrong. Should also expose TLS library linkage
542 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
543 64-bit ${eval} (JH/03).
545 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
546 GNU libc to support some of the 64-bit stuff, should not lead to
547 conflicts. Defined before os.h is pulled in, so if a given platform
548 needs to override this, it can.
550 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
551 protection layer was required, which is not implemented.
552 Bugzilla 1254, patch from Wolfgang Breyha.
554 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
555 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
556 tls_dhparam take prime identifiers. Also unbreak combination of
557 OpenSSL+DH_params+TLSSNI.
559 PP/39 Disable SSLv2 by default in OpenSSL support.
565 PP/01 Solaris build fix for Oracle's LDAP libraries.
566 Bugzilla 1109, patch from Stephen Usher.
568 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
570 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
573 TF/02 Fix a couple more cases where we did not log the error message
574 when unlink() failed. See also change 4.74-TF/03.
576 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
577 lock up or crash if it happened to be inside a call to libc when it
578 got a SIGUSR1 from exiwhat.
580 The SIGUSR1 handler appends the current process status to the process
581 log which is later printed by exiwhat. It used to use the general
582 purpose logging code to do this, but several functions it calls are
583 not safe for signals.
585 The new output code in the SIGUSR1 handler is specific to the process
586 log, and simple enough that it's easy to inspect for signal safety.
587 Removing some special cases also simplifies the general logging code.
588 Removing the spurious timestamps from the process log simplifies
591 TF/04 Improved ratelimit ACL condition.
593 The /noupdate option has been deprecated in favour of /readonly which
594 has clearer semantics. The /leaky, /strict, and /readonly update modes
595 are mutually exclusive. The update mode is no longer included in the
596 database key; it just determines when the database is updated. (This
597 means that when you upgrde Exim will forget old rate measurements.)
599 Exim now checks that the per_* options are used with an update mode that
600 makes sense for the current ACL. For example, when Exim is processing a
601 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
602 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
603 must specify per_mail/readonly. If you omit the update mode it defaults to
604 /leaky where that makes sense (as before) or /readonly where required.
606 The /noupdate option is now undocumented but still supported for
607 backwards compatibility. It is equivalent to /readonly except that in
608 ACLs where /readonly is required you may specify /leaky/noupdate or
609 /strict/noupdate which are treated the same as /readonly.
611 A useful new feature is the /count= option. This is a generalization
612 of the per_byte option, so that you can measure the throughput of other
613 aggregate values. For example, the per_byte option is now equivalent
614 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
616 The per_rcpt option has been generalized using the /count= mechanism
617 (though it's more complicated than the per_byte equivalence). When it is
618 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
619 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
620 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
621 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
622 Note that using per_rcpt with a non-readonly update mode in more than
623 one ACL will cause the recipients to be double-counted. (The per_mail
624 and per_byte options don't have this problem.)
626 The handling of very low rates has changed slightly. If the computed rate
627 is less than the event's count (usually one) then this event is the first
628 after a long gap. In this case the rate is set to the same as this event's
629 count, so that the first message of a spam run is counted properly.
631 The major new feature is a mechanism for counting the rate of unique
632 events. The new per_addr option counts the number of different
633 recipients that someone has sent messages to in the last time period. It
634 behaves like per_rcpt if all the recipient addresses are different, but
635 duplicate recipient addresses do not increase the measured rate. Like
636 the /count= option this is a general mechanism, so the per_addr option
637 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
638 example, measure the rate that a client uses different sender addresses
639 with the options per_mail/unique=$sender_address. There are further
640 details in the main documentation.
642 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
644 TF/06 Removed a few PCRE remnants.
646 TF/07 Automatically extract Exim's version number from tags in the git
647 repository when doing development or release builds.
649 PP/02 Raise smtp_cmd_buffer_size to 16kB.
650 Bugzilla 879. Patch from Paul Fisher.
652 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
653 Heavily based on revision 40f9a89a from Simon Arlott's tree.
656 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
658 PP/05 Variable $av_failed, true if the AV scanner deferred.
659 Bugzilla 1078. Patch from John Horne.
661 PP/06 Stop make process more reliably on build failure.
662 Bugzilla 1087. Patch from Heiko Schlittermann.
664 PP/07 Make maildir_use_size_file an _expandable_ boolean.
665 Bugzilla 1089. Patch from Heiko Schlittermann.
667 PP/08 Handle ${run} returning more data than OS pipe buffer size.
668 Bugzilla 1131. Patch from Holger Weiß.
670 PP/09 Handle IPv6 addresses with SPF.
671 Bugzilla 860. Patch from Wolfgang Breyha.
673 PP/10 GnuTLS: support TLS 1.2 & 1.1.
675 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
678 PP/11 match_* no longer expand right-hand-side by default.
679 New compile-time build option, EXPAND_LISTMATCH_RHS.
680 New expansion conditions, "inlist", "inlisti".
682 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
684 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
686 PP/14 fix log_write() format string regression from TF/03.
687 Bugzilla 1152. Patch from Dmitry Isaikin.
693 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
695 PP/02 Harmonised TLS library version reporting; only show if debugging.
696 Layout now matches that introduced for other libraries in 4.74 PP/03.
698 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
700 PP/04 New "dns_use_edns0" global option.
702 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
705 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
706 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
708 TK/01 Updated PolarSSL code to 0.14.2.
709 Bugzilla 1097. Patch from Andreas Metzler.
711 PP/07 Catch divide-by-zero in ${eval:...}.
714 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
717 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
718 format-string attack -- SECURITY: remote arbitrary code execution.
720 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
721 time unintentionally subject to list matching rules, letting the header
722 cause arbitrary Exim lookups (of items which can occur in lists, *not*
723 arbitrary string expansion). This allowed for information disclosure.
725 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
726 INT_MIN/-1 -- value coerced to INT_MAX.
732 NM/01 Workround for PCRE version dependency in version reporting
735 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
736 This fixes portability to compilers other than gcc, notably
737 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
739 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
740 makefiles for portability to HP-UX and POSIX correctness.
742 PP/01 Permit LOOKUP_foo enabling on the make command-line.
743 Also via indented variable definition in the Makefile.
744 (Debugging by Oliver Heesakkers).
746 PP/02 Restore caching of spamd results with expanded spamd_address.
747 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
749 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
750 Improves build reliability. Fix from: Frank Elsner
752 NM/02 Fix wide character breakage in the rfc2047 coding
753 Fixes bug 1064. Patch from Andrey N. Oktyabrski
755 NM/03 Allow underscore in dnslist lookups
756 Fixes bug 1026. Patch from Graeme Fowler
758 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
759 Code patches from Adam Ciarcinski of NetBSD.
761 NM/04 Fixed exiqgrep to cope with mailq missing size issue
764 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
765 is logged, to avoid truncation. Patch from John Horne.
767 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
768 Patch from Jakob Hirsch.
770 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
771 SQL string expansion failure details.
772 Patch from Andrey Oktyabrski.
774 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
775 Patch from Simon Arlott.
777 PP/09 New lookups functionality failed to compile on old gcc which rejects
778 extern declarations in function scope.
779 Patch from Oliver Fleischmann
781 PP/10 Use sig_atomic_t for flags set from signal handlers.
782 Check getgroups() return and improve debugging.
783 Fixed developed for diagnosis in bug 927 (which turned out to be
786 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
787 Patch from Mark Zealey.
789 PP/12 Bugzilla 1056: Improved spamd server selection.
790 Patch from Mark Zealey.
792 PP/13 Bugzilla 1086: Deal with maildir quota file races.
793 Based on patch from Heiko Schlittermann.
795 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
796 Patch from Uwe Doering, sign-off by Michael Haardt.
798 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
799 variable declaration deep within a block. Bug and patch from
802 PP/15 lookups-Makefile IRIX compatibilty coercion.
804 PP/16 Make DISABLE_DKIM build knob functional.
806 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
807 Patch by Simon Arlott
809 TF/03 Fix valgrind.h portability to C89 compilers that do not support
810 variable argument macros. Our copy now differs from upstream.
816 TF/01 Failure to get a lock on a hints database can have serious
817 consequences so log it to the panic log.
819 TF/02 Log LMTP confirmation messages in the same way as SMTP,
820 controlled using the smtp_confirmation log selector.
822 TF/03 Include the error message when we fail to unlink a spool file.
824 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
825 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
826 for maintaining out-of-tree patches for some time.
828 PP/01 Bugzilla 139: Documentation and portability issues.
829 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
830 Handle per-OS dynamic-module compilation flags.
832 PP/02 Let /dev/null have normal permissions.
833 The 4.73 fixes were a little too stringent and complained about the
834 permissions on /dev/null. Exempt it from some checks.
835 Reported by Andreas M. Kirchwitz.
837 PP/03 Report version information for many libraries, including
838 Exim version information for dynamically loaded libraries. Created
839 version.h, now support a version extension string for distributors
840 who patch heavily. Dynamic module ABI change.
842 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
843 privilege escalation vulnerability whereby the Exim run-time user
844 can cause root to append content of the attacker's choosing to
847 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
850 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
851 If dropping privileges for untrusted macros, we disabled normal logging
852 on the basis that it would fail; for the Exim run-time user, this is not
853 the case, and it resulted in successful deliveries going unlogged.
854 Fixed. Reported by Andreas Metzler.
860 PP/01 Date: & Message-Id: revert to normally being appended to a message,
861 only prepend for the Resent-* case. Fixes regression introduced in
862 Exim 4.70 by NM/22 for Bugzilla 607.
864 PP/02 Include check_rfc2047_length in configure.default because we're seeing
865 increasing numbers of administrators be bitten by this.
867 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
869 PP/03 Bugzilla 994: added openssl_options main configuration option.
871 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
873 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
875 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
877 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
878 without a peer certificate, leading to a segfault because of an
879 assumption that peers always have certificates. Be a little more
880 paranoid. Problem reported by Martin Tscholak.
882 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
883 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
884 NB: ClamAV planning to remove STREAM in "middle of 2010".
885 CL also introduces -bmalware, various -d+acl logging additions and
886 more caution in buffer sizes.
888 PP/09 Implemented reverse_ip expansion operator.
890 PP/10 Bugzilla 937: provide a "debug" ACL control.
892 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
894 PP/12 Bugzilla 973: Implement --version.
896 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
898 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
900 PP/15 Bugzilla 816: support multiple condition rules on Routers.
902 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
903 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
904 ignore trailing whitespace.
906 JJ/02 prevent non-panic DKIM error from being sent to paniclog
908 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
911 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
912 Notification from Dr Andrew Aitchison.
914 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
915 ExtendedDetectionInfo response format.
916 Notification from John Horne.
918 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
921 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
922 XSL and documented dependency on system catalogs, with examples of how
925 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
928 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
929 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
930 configuration file which is writeable by the Exim user or group.
932 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
933 of configuration files to cover files specified with the -C option if
934 they are going to be used with root privileges, not just the default
937 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
938 option (effectively making it always true).
940 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
941 files to be used while preserving root privileges.
943 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
944 that rogue child processes cannot use them.
946 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
947 run-time user, instead of root.
949 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
950 Exim run-time user without dropping privileges.
952 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
953 result string, instead of calling string_vformat() twice with the same
956 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
957 for other users. Others should always drop root privileges if they use
958 -C on the command line, even for a whitelisted configure file.
960 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
962 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
968 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
969 $data_path, and $header_path variables; fixed documentation bugs and
972 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
973 exipick to access non-standard spools, including the "frozen" queue
976 NM/01 Bugzilla 965: Support mysql stored procedures.
977 Patch from Alain Williams
979 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
981 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
982 Patch from Andreas Metzler
984 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
985 Patch from Kirill Miazine
987 NM/05 Bugzilla 671: Added umask to procmail example.
989 JJ/03 installed exipick 20100323.0, fixing doc bug
991 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
992 directory. Notification and patch from Dan Rosenberg.
994 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
996 TK/02 Improve log output when DKIM signing operation fails.
998 MH/01 Treat the transport option dkim_domain as a colon separated
999 list, not as a single string, and sign the message with each element,
1000 omitting multiple occurences of the same signer.
1002 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1003 Bugzilla 985, 986. Patch by Simon Arlott
1005 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1006 Patch by Simon Arlott
1008 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1009 MBX locking. Notification from Dan Rosenberg.
1015 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1017 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1019 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1021 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1023 NM/04 Bugzilla 915: Flag broken perl installation during build.
1029 TK/01 Added patch by Johannes Berg that expands the main option
1030 "spamd_address" if it starts with a dollar sign.
1032 TK/02 Write list of recipients to X-Envelope-Sender header when building
1033 the mbox-format spool file for content scanning (suggested by Jakob
1036 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1037 (http://www.dcc-servers.net/) support via dccifd. Activated by
1038 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1040 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1041 by Mark Daniel Reidel <mr@df.eu>.
1043 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1044 When building exim an external PCRE library is now needed -
1045 PCRE is a system library on the majority of modern systems.
1046 See entry on PCRE_LIBS in EDITME file.
1048 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1049 conversation. Added nologin parameter to request.
1050 Patch contributed by Kirill Miazine.
1052 TF/01 Do not log submission mode rewrites if they do not change the address.
1054 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1056 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1057 log files in place. Contributed by Roberto Lima.
1059 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1061 TF/03 Bugzilla 615: When checking the local_parts router precondition
1062 after a local_part_suffix or local_part_prefix option, Exim now
1063 does not use the address's named list lookup cache, since this
1064 contains cached lookups for the whole local part.
1066 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1067 Robert Millan. Documentation is in experimental-spec.txt.
1069 TF/04 Bugzilla 668: Fix parallel build (make -j).
1071 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1073 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1074 Patch provided by Jan Srzednicki.
1076 TF/05 Leading white space used to be stripped from $spam_report which
1077 wrecked the formatting. Now it is preserved.
1079 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1080 that they are available at delivery time.
1082 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1084 TF/08 TLS error reporting now respects the incoming_interface and
1085 incoming_port log selectors.
1087 TF/09 Produce a more useful error message if an SMTP transport's hosts
1088 setting expands to an empty string.
1090 NM/06 Bugzilla 744: EXPN did not work under TLS.
1091 Patch provided by Phil Pennock.
1093 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1094 Patch provided by Richard Godbee.
1096 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1097 acl_smtp_notquit, added index entry.
1099 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1100 Patch provided by Eugene Bujak.
1102 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1103 accept(). Patch provided by Maxim Dounin.
1105 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1106 Patch provided by Phil Pennock.
1108 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1110 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1111 Patch provided by Brad "anomie" Jorsch.
1113 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1114 Patch provided by Dean Brooks.
1116 TK/05 Add native DKIM support (does not depend on external libraries).
1118 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1119 Patch provided by Graeme Fowler.
1121 NM/16 Bugzilla 851: Documentation example syntax fix.
1123 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1125 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1128 NM/19 Bugzilla 745: TLS version reporting.
1129 Patch provided by Phil Pennock.
1131 NM/20 Bugzilla 167: bool: condition support.
1132 Patch provided by Phil Pennock.
1134 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1135 clients. Patch provided by Phil Pennock.
1137 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1138 Patch provided by Brad "anomie" Jorsch.
1140 NM/23 Bugzilla 687: Fix misparses in eximstats.
1141 Patch provided by Heiko Schlittermann.
1143 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1144 Patch provided by Heiko Schlittermann.
1146 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1147 plus update to original patch.
1149 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1151 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1152 Patch provided by David Brownlee.
1154 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1156 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1158 NM/30 Bugzilla 888: TLS documentation bugfixes.
1160 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1162 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1163 Unlike the original bugzilla I have changed all shell scripts in src tree.
1165 NM/33 Bugzilla 898: Transport filter timeout fix.
1166 Patch by Todd Rinaldo.
1168 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1169 Patch by Serge Demonchaux.
1171 NM/35 Bugzilla 39: Base64 decode bug fixes.
1172 Patch by Jakob Hirsch.
1174 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1176 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1178 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1180 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1186 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1187 ALT-N's libdkim that I have put here:
1188 http://duncanthrax.net/exim-experimental/
1190 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1191 were called 'true' and it seems that C99 defines that as a reserved
1192 keyword to be used with 'bool' variable types. That means you could
1193 not include C99-style headers which use bools without triggering
1194 build errors in sieve.c.
1196 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1197 as mailq or other aliases. Changed the --help handling significantly
1198 to do whats expected. exim_usage() emits usage/help information.
1200 SC/01 Added the -bylocaldomain option to eximstats.
1202 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1204 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1206 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1212 PH/01 Another patch from the Sieve maintainer.
1214 PH/02 When an IPv6 address is converted to a string for single-key lookup
1215 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1216 dots are used instead of colons so that keys in lsearch files need not
1217 contain colons. This was done some time before quoting was made available
1218 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1219 (notated using the quote facility) so as to distinguish them from IPv4
1220 keys. This meant that lookups for IP addresses in host lists did not work
1221 for iplsearch lookups.
1223 This has been fixed by arranging for IPv6 addresses to be expressed with
1224 colons if the lookup type is iplsearch. This is not incompatible, because
1225 previously such lookups could never work.
1227 The situation is now rather anomolous, since one *can* have colons in
1228 ordinary lsearch keys. However, making the change in all cases is
1229 incompatible and would probably break a number of configurations.
1231 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1234 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1235 conversion specification without a maximum field width, thereby enabling
1236 a rogue spamd server to cause a buffer overflow. While nobody in their
1237 right mind would setup Exim to query an untrusted spamd server, an
1238 attacker that gains access to a server running spamd could potentially
1239 exploit this vulnerability to run arbitrary code as the Exim user.
1241 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1242 $primary_hostname instead of what libspf2 thinks the hosts name is.
1244 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1245 a directory entry by the name of the lookup key. Previously, if a
1246 symlink pointed to a non-existing file or a file in a directory that
1247 Exim lacked permissions to read, a lookup for a key matching that
1248 symlink would fail. Now it is enough that a matching directory entry
1249 exists, symlink or not. (Bugzilla 503.)
1251 PH/03 The body_linecount and body_zerocount variables are now exported in the
1254 PH/04 Added the $dnslist_matched variable.
1256 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1257 This means they are set thereafter only if the connection becomes
1260 PH/06 Added the client_condition to authenticators so that some can be skipped
1261 by clients under certain conditions.
1263 PH/07 The error message for a badly-placed control=no_multiline_responses left
1264 "_responses" off the end of the name.
1266 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1268 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1269 (without spaces) instead of just copying the configuration text.
1271 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1273 PH/11 Added $max_received_linelength.
1275 PH/12 Added +ignore_defer and +include_defer to host lists.
1277 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1278 way in which PCRE > 7.0 is built.
1280 PH/14 Implemented queue_only_load_latch.
1282 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1283 MAIL command. The effect was to mangle the value on 64-bit systems.
1285 PH/16 Another patch from the Sieve maintainer.
1287 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1289 PH/18 If a system quota error occurred while trying to create the file for
1290 a maildir delivery, the message "Mailbox is full" was not appended to the
1291 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1292 only to a quota excession during the actual writing of the file.
1294 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1295 characters?) which causes problems in log lines. The DN values are now
1296 passed through string_printing() before being added to log lines.
1298 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1299 and InterBase are left for another time.)
1301 PH/21 Added message_body_newlines option.
1303 PH/22 Guard against possible overflow in moan_check_errorcopy().
1305 PH/23 POSIX allows open() to be a macro; guard against that.
1307 PH/24 If the recipient of an error message contained an @ in the local part
1308 (suitably quoted, of course), incorrect values were put in $domain and
1309 $local_part during the evaluation of errors_copy.
1315 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1316 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1319 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1320 issue a MAIL command.
1322 PH/02 In an ACL statement such as
1324 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1326 if a client was not listed at all, or was listed with a value other than
1327 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1328 the condition was not true (as it should be), so access was not denied.
1329 The bug was that the ! inversion was incorrectly passed on to the second
1330 item. This has been fixed.
1332 PH/03 Added additional dnslists conditions == and =& which are different from
1333 = and & when the dns lookup returns more than one IP address.
1335 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1336 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1338 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1339 FSYNC, which compiles an option called disable_fsync that allows for
1340 bypassing fsync(). The documentation is heavily laced with warnings.
1342 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1344 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1345 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1346 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1347 including adding "make clean"; (3) Added -fPIC when compiling the test
1348 dynamically loaded module, to get rid of a warning.
1350 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1351 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1352 = 0s. The bug is otherwise harmless.
1354 PH/07 There was a bug in the dovecot authenticator such that the value of
1355 $auth1 could be overwritten, and so not correctly preserved, after a
1356 successful authentication. This usually meant that the value preserved by
1357 the server_setid option was incorrect.
1359 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1361 PH/09 Installed PCRE release 7.0.
1363 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1364 run for batched SMTP input. It is now run at the start of every message
1365 in the batch. While fixing this I discovered that the process information
1366 (output by running exiwhat) was not always getting set for -bs and -bS
1367 input. This is fixed, and it now also says "batched" for BSMTP.
1369 PH/11 Added control=no_pipelining.
1371 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1372 patch, slightly modified), and move the expansion of helo_data till after
1373 the connection is made in the smtp transport (so it can use these
1376 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1378 PH/14 Added log_selector = +pid.
1380 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1382 PH/16 Add ${if forany and ${if forall.
1384 PH/17 Added dsn_from option to vary the From: line in DSNs.
1386 PH/18 Flush SMTP output before performing a callout, unless control =
1387 no_callout_flush is set.
1389 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1390 was true (the default) a successful delivery failed to delete the retry
1391 item, thus causing premature timeout of the address. The bug is now
1394 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1396 PH/21 Long custom messages for fakedefer and fakereject are now split up
1397 into multiline reponses in the same way that messages for "deny" and
1398 other ACL rejections are.
1400 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1401 with slight modification.
1403 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1404 draft, changing the syntax and factoring some duplicate code".
1406 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1407 for deliveries of the second and subsequent messages over the same SMTP
1410 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1411 ${reduce, with only minor "tidies".
1413 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1415 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1416 expansion side effects.
1418 PH/27 When a message times out after an over-quota error from an Exim-imposed
1419 quota, the bounce message says "mailbox is full". This message was not
1420 being given when it was a system quota that was exceeded. It now should
1423 MH/03 Made $recipients available in local_scan(). local_scan() already has
1424 better access to the recipient list through recipients_list[], but
1425 $recipients can be useful in postmaster-provided expansion strings.
1427 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1428 in the case of a MAIL command with additional options following the
1429 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1430 were accidentally chopped off.
1432 PH/29 SMTP synchronization checks are implemented when a command is read -
1433 there is a check that no more input is waiting when there shouldn't be
1434 any. However, for some commands, a delay in an ACL can mean that it is
1435 some time before the response is written. In this time, more input might
1436 arrive, invalidly. So now there are extra checks after an ACL has run for
1437 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1438 pipelining has not been advertised.
1440 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1442 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1443 returned in the SMTP response when smtp_return_error_details was false.
1444 This has been fixed.
1446 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1447 instead of the C I/O that was originally supplied, because problems were
1448 reported on Solaris.
1450 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1451 Exim which did not show up earlier: it was assuming that a call to
1452 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1453 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1454 was a macro that became an assignment, so it seemed to work. This has
1455 changed to a proper function call with a void return, hence the compile
1456 error. Exim's code has been fixed.
1458 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1461 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1463 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1465 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1466 because it turns out to be too big or there is a timeout) there is no
1467 "Completed" line in the log. When some messages of this type were
1468 selected by exigrep, they were listed as "not completed". Others were
1469 picked up by some special patterns. I have improved the selection
1470 criteria to be more general.
1472 PH/38 The host_find_failed option in the manualroute router can now be set
1473 to "ignore", to completely ignore a host whose IP address cannot be
1474 found. If all hosts are ignored, the behaviour is controlled by the new
1475 host_all_ignored option.
1477 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1478 homing or because of multiple MX records with /mx) generated more than
1479 one IP address, and the following item turned out to be the local host,
1480 all the secondary addresses of the first item were incorrectly removed
1481 from the list, along with the local host and any following hosts (which
1482 is what is supposed to happen).
1484 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1485 whoever called Exim into the -H file. In the case of the daemon it was
1486 behaving confusingly. When first started, it used values for whoever
1487 started the daemon, but after a SIGHUP it used the Exim user (because it
1488 calls itself on a restart). I have changed the code so that it now always
1491 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1492 message are rejected with the same error (e.g. no authentication or bad
1493 sender address), and a DATA command is nevertheless sent (as can happen
1494 with PIPELINING or a stupid MUA), the error message that was given to the
1495 RCPT commands is included in the rejection of the DATA command. This is
1496 intended to be helpful for MUAs that show only the final error to their
1499 PH/42 Another patch from the Sieve maintainer.
1501 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1502 Eximstats - Fixed some broken HTML links and added missing column headers
1504 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1505 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1507 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1509 PH/43 Yet another patch from the Sieve maintainer.
1511 PH/44 I found a way to check for a TCP/IP connection going away before sending
1512 the response to the final '.' that terminates a message, but only in the
1513 case where the client has not sent further data following the '.'
1514 (unfortunately, this is allowed). However, in many cases there won't be
1515 any further data because there won't be any more messages to send. A call
1516 to select() can be used: if it shows that the input is "ready", there is
1517 either input waiting, or the socket has been closed. An attempt to read
1518 the next input character can distinguish the two cases. Previously, Exim
1519 would have sent an OK response which the client would never have see.
1520 This could lead to message repetition. This fix should cure that, at
1521 least in a lot of common cases.
1523 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1524 advertised in response to EHLO.
1530 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1531 fixed by 4.65/MH/01 (is this a record?) are fixed:
1533 (i) An empty string was always treated as zero by the numeric comparison
1534 operators. This behaviour has been restored.
1536 (ii) It is documented that the numeric comparison operators always treat
1537 their arguments as decimal numbers. This was broken in that numbers
1538 starting with 0 were being interpreted as octal.
1540 While fixing these problems I realized that there was another issue that
1541 hadn't been noticed. Values of message_size_limit (both the global option
1542 and the transport option) were treated as octal if they started with 0.
1543 The documentation was vague. These values are now always treated as
1544 decimal, and I will make that clear in the documentation.
1550 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1551 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1554 MH/01 Don't check that the operands of numeric comparison operators are
1555 integers when their expansion is in "skipping" mode (fixes bug
1556 introduced by 4.64-PH/07).
1558 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1559 child addresses, Exim now panics and dies. Previously, because the count
1560 is held in a short int, deliveries were likely to be lost. As such a
1561 large number of recipients for a single message is ridiculous
1562 (performance will be very, very poor), I have chosen to impose a limit
1563 rather than extend the field.
1569 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1570 leftover -K file (the existence of which was triggered by #402).
1571 While we were at it, introduced process PID as part of the -K
1572 filename. This should rule out race conditions when creating
1575 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1576 processing considerably. Previous code took too long for large mails,
1577 triggering a timeout which in turn triggers #401.
1579 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1580 in the DK code in transports.c. sendfile() is not really portable,
1581 hence the _LINUX specificness.
1583 TF/01 In the add_headers option to the mail command in an Exim filter,
1584 there was a bug that Exim would claim a syntax error in any
1585 header after the first one which had an odd number of characters
1588 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1589 callout verification, Exim cached a "reject" for the entire domain. This
1590 is correct for most verifications, but it is not correct for a recipient
1591 verification with use_sender or use_postmaster set, because in that case
1592 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1593 case of MAIL FROM:<> rejection from other early rejections (e.g.
1594 rejection of HELO). When verifying a recipient using a non-null MAIL
1595 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1596 Whatever the result of the callout, the value of the domain cache is
1597 left unchanged (for any other kind of callout, getting as far as trying
1598 RCPT means that the domain itself is ok).
1600 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
1603 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
1604 manifest itself as EPIPE rather than ECONNECT. When tidying away a
1605 session, the daemon ignores ECONNECT errors and logs others; it now
1606 ignores EPIPE as well.
1608 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
1609 (quoted-printable decoding).
1611 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
1612 later the small subsequent patch to fix an introduced bug.
1614 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
1616 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
1618 PH/08 An error is now given if message_size_limit is specified negative.
1620 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
1621 to be given (somewhat) arbitrary names.
1623 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
1626 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
1627 miscellaneous code fixes
1629 PH/10 Added the log_reject_target ACL modifier to specify where to log
1632 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
1633 hostname. This is wrong, because it relates to the incoming message (and
1634 probably the interface on which it is arriving) and not to the outgoing
1635 callout (which could be using a different interface). This has been
1636 changed to use the value of the helo_data option from the smtp transport
1637 instead - this is what is used when a message is actually being sent. If
1638 there is no remote transport (possible with a router that sets up host
1639 addresses), $smtp_active_hostname is used.
1641 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
1642 tweaks were necessary in order to get it to work (see also 21 below):
1643 (a) The code assumed that strncpy() returns a negative number on buffer
1644 overflow, which isn't the case. Replaced with Exim's string_format()
1646 (b) There were several signed/unsigned issues. I just did the minimum
1647 hacking in of casts. There is scope for a larger refactoring.
1648 (c) The code used strcasecmp() which is not a standard C function.
1649 Replaced with Exim's strcmpic() function.
1650 (d) The code set only $1; it now sets $auth1 as well.
1651 (e) A simple test gave the error "authentication client didn't specify
1652 service in request". It would seem that Dovecot has changed its
1653 interface. Fortunately there's a specification; I followed it and
1654 changed what the client sends and it appears to be working now.
1656 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
1659 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
1660 address A is aliased to B and C, where B exists and C does not. Without
1661 -v the output is "A verified" because verification stops after a
1662 successful redirection if more than one address is generated. However,
1663 with -v the child addresses are also verified. Exim was outputting "A
1664 failed to verify" and then showing the successful verification for C,
1665 with its parentage. It now outputs "B failed to verify", showing B's
1666 parentage before showing the successful verification of C.
1668 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
1669 look up a TXT record in a specific list after matching in a combined
1672 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
1673 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
1674 they consult the DNS. I had assumed they would set it the way they
1675 wanted; and indeed my experiments on Linux seem to show that in some
1676 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
1677 To be on the safe side, however, I have now made the interface to
1678 host_find_byname() similar to host_find_bydns(), with an argument
1679 containing the DNS resolver options. The host_find_byname() function now
1680 sets these options at its start, just as host_find_bydns() does. The smtp
1681 transport options dns_qualify_single and dns_search_parents are passed to
1682 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
1683 of host_find_byname() use the default settings of RES_DEFNAMES
1684 (qualify_single) but not RES_DNSRCH (search_parents).
1686 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
1687 spool_read_header() do less string testing, by means of a preliminary
1688 switch on the second character of optional "-foo" lines. (This is
1689 overdue, caused by the large number of possibilities that now exist.
1690 Originally there were few.) While I was there, I also converted the
1691 str(n)cmp tests so they don't re-test the leading "-" and the first
1692 character, in the hope this might squeeze out yet more improvement.
1694 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
1695 flag allowing group syntax was set by the header_syntax check but not
1696 turned off, possible causing trouble later; (2) The flag was not being
1697 set at all for the header_verify test, causing "group"-style headers to
1698 be rejected. I have now set it in this case, and also caused header_
1699 verify to ignore an empty address taken from a group. While doing this, I
1700 came across some other cases where the code for allowing group syntax
1701 while scanning a header line wasn't quite right (mostly, not resetting
1702 the flag correctly in the right place). These bugs could have caused
1703 trouble for malformed header lines. I hope it is now all correct.
1705 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
1706 with the "reply" argument non-NULL. The code, however (which originally
1707 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
1708 but it didn't always do it. This confused somebody who was copying the
1709 code for some other use. I have removed all the tests.
1711 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
1712 feature that was used to support insecure browsers during the U.S. crypto
1713 embargo. It requires special client support, and Exim is probably the
1714 only MTA that supported it -- and would never use it because real RSA is
1715 always available. This code has been removed, because it had the bad
1716 effect of slowing Exim down by computing (never used) parameters for the
1717 RSA_EXPORT functionality.
1719 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
1720 authenticator to fail if there's a tab character in the incoming data
1721 (there should never be unless someone is messing about, as it's supposed
1722 to be base64-encoded). Also added, on Timo's advice, the "secured" option
1723 if the connection is using TLS or if the remote IP is the same as the
1724 local IP, and the "valid-client-cert option" if a client certificate has
1727 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
1728 authenticators. This can be used for authorization after authentication
1729 succeeds. (In the case of plaintext, it servers for both authentication
1732 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
1733 if any retry times were supplied.
1735 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
1736 connection, where there is no client IP address to check. In this
1737 situation, the verify now always succeeds.
1739 PH/25 Applied John Jetmore's -Mset patch.
1741 PH/26 Added -bem to be like -Mset, but loading a message from a file.
1743 PH/27 In a string expansion for a processed (not raw) header when multiple
1744 headers of the same name were present, leading whitespace was being
1745 removed from all of them, but trailing whitespace was being removed only
1746 from the last one. Now trailing whitespace is removed from each header
1747 before concatenation. Completely empty headers in a concatenation (as
1748 before) are ignored.
1750 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
1751 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
1753 PH/29 [Removed. This was a change that I later backed out, and forgot to
1754 correct the ChangeLog entry (that I had efficiently created) before
1755 committing the later change.]
1757 PH/30 Exim was sometimes attempting to deliver messages that had suffered
1758 address errors (4xx response to RCPT) over the same connection as other
1759 messages routed to the same hosts. Such deliveries are always "forced",
1760 so retry times are not inspected. This resulted in far too many retries
1761 for the affected addresses. The effect occurred only when there were more
1762 hosts than the hosts_max_try setting in the smtp transport when it had
1763 the 4xx errors. Those hosts that it had tried were not added to the list
1764 of hosts for which the message was waiting, so if all were tried, there
1765 was no problem. Two fixes have been applied:
1767 (i) If there are any address or message errors in an SMTP delivery, none
1768 of the hosts (tried or untried) are now added to the list of hosts
1769 for which the message is waiting, so the message should not be a
1770 candidate for sending over the same connection that was used for a
1771 successful delivery of some other message. This seems entirely
1772 reasonable: after all the message is NOT "waiting for some host".
1773 This is so "obvious" that I'm not sure why it wasn't done
1774 previously. Hope I haven't missed anything, but it can't do any
1775 harm, as the worst effect is to miss an optimization.
1777 (ii) If, despite (i), such a delivery is accidentally attempted, the
1778 routing retry time is respected, so at least it doesn't keep
1779 hammering the server.
1781 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
1782 in ${readsocket because some servers need this prod.
1784 PH/32 Added some extra debug output when updating a wait-xxx database.
1786 PH/33 The hint "could be header name not terminated by colon", which has been
1787 given for certain expansion errors for a long time, was not being given
1788 for the ${if def:h_colon_omitted{... case.
1790 PH/34 The spec says: "With one important exception, whenever a domain list is
1791 being scanned, $domain contains the subject domain." There was at least
1792 one case where this was not true.
1794 PH/35 The error "getsockname() failed: connection reset by peer" was being
1795 written to the panic log as well as the main log, but it isn't really
1796 panic-worthy as it just means the connection died rather early on. I have
1797 removed the panic log writing for the ECONNRESET error when getsockname()
1800 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
1801 runs only) independently of the message's sender address. This meant
1802 that, if the 4xx error was in fact related to the sender, a different
1803 message to the same recipient with a different sender could confuse
1804 things. In particualar, this can happen when sending to a greylisting
1805 server, but other circumstances could also provoke similar problems.
1806 I have changed the default so that the retry time for these errors is now
1807 based a combination of the sender and recipient addresses. This change
1808 can be overridden by setting address_retry_include_sender=false in the
1811 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
1812 remote server are returned as part of bounce messages. This was not
1813 happening for LMTP over a pipe (the lmtp transport), but now it is the
1814 same for both kinds of LMTP.
1816 PH/38 Despite being documented as not happening, Exim was rewriting addresses
1817 in header lines that were in fact CNAMEs. This is no longer the case.
1819 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
1820 and queue runs started by the daemon processed all messages. This has
1821 been fixed so that -R and -S can now usefully be given with -q<time>.
1823 PH/40 Import PCRE release 6.7 (fixes some bugs).
1825 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
1827 PH/42 Give an error if -q is specified more than once.
1829 PH/43 Renamed the variables $interface_address and $interface_port as
1830 $received_ip_address and $received_port, to make it clear that these
1831 values apply to message reception, and not to the outgoing interface when
1832 a message is delivered. (The old names remain recognized, of course.)
1834 PH/44 There was no timeout on the connect() call when using a Unix domain
1835 socket in the ${readsocket expansion. There now is.
1837 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
1838 be meaningful with "accept".
1840 SC/01 Eximstats V1.43
1841 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
1843 SC/02 Eximstats V1.44
1844 Use a glob alias rather than an array ref in the generated
1845 parser. This improves both readability and performance.
1847 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
1848 Collect SpamAssassin and rejection statistics.
1849 Don't display local sender or destination tables unless
1850 there is data to show.
1851 Added average volumes into the top table text output.
1853 SC/04 Eximstats V1.46
1854 Collect data on the number of addresses (recipients)
1855 as well as the number of messages.
1857 SC/05 Eximstats V1.47
1858 Added 'Message too big' to the list of mail rejection
1859 reasons (thanks to Marco Gaiarin).
1861 SC/06 Eximstats V1.48
1862 Mainlog lines which have GMT offsets and are too short to
1863 have a flag are now skipped.
1865 SC/07 Eximstats V1.49 (Alain Williams)
1866 Added the -emptyok flag.
1868 SC/08 Eximstats V1.50
1869 Fixes for obtaining the IP address from reject messages.
1871 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
1872 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
1873 whitesspace changes from 4.64-PH/27
1875 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
1878 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
1879 are found, allow negative numbers in numeric criteria)
1881 JJ/06 exipick.20061117.2, added new $message_body_missing variable
1883 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
1884 to match changes made in 4.64-PH/43
1886 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
1888 PH/47 Put in an explicit test for a DNS lookup of an address record where the
1889 "domain" is actually an IP address, and force a failure. This locks out
1890 those revolvers/nameservers that support "A-for-A" lookups, in
1891 contravention of the specifications.
1893 PH/48 When a host name was looked up from an IP address, and the subsequent
1894 forward lookup of the name timed out, the host name was left in
1895 $sender_host_name, contrary to the specification.
1897 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
1898 restricted to single-key lookups, Exim was not diagnosing an error if
1899 * or *@ was used with a query-style lookup.
1901 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
1903 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
1904 long ago, but noone interested enough thought of it. Let's just say that
1905 the "1.1" means that there are some new functions that weren't there at
1906 some point in the past.
1908 PH/51 Error processing for expansion failure of helo_data from an smtp
1909 transport during callout processing was broken.
1911 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
1912 tested/used via the -bh/-bhc/-bs options.
1914 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
1915 bug, fixed in subsequent PCRE releases).
1917 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
1918 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
1920 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
1926 SC/01 Use a glob alias rather than an array ref in eximstats generated
1927 parser. This improves both readability and performance.
1929 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
1930 Don't display local sender or destination tables in eximstats unless
1931 there is data to show.
1932 Added average volumes into the eximstats top table text output.
1934 SC/03 Collect data on the number of addresses (recipients) as well
1935 as the number of messages in eximstats.
1937 TF/01 Correct an error in the documentation for the redirect router. Exim
1938 does (usually) call initgroups() when daemonizing.
1940 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
1941 with consistent privilege compared to when running as a daemon.
1943 TF/03 Note in the spec that $authenticated_id is not set for local
1944 submissions from trusted users.
1946 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
1947 Thanks to Dean Brooks <dean@iglou.com> for the patch.
1949 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
1950 by adding some example configuration directives to the default
1951 configuration file. A little bit of work is required to uncomment the
1952 directives and define how usernames and passwords are checked, but
1953 there is now a framework to start from.
1955 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
1956 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
1957 without this. I don't know how relevant this is to other LDAP libraries.
1959 PH/02 Add the verb name to the "unknown ACL verb" error.
1961 PH/03 Magnus Holmgren's patch for filter_prepend_home.
1963 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
1965 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
1966 directory not expanded when it should be if an expanded home directory
1967 was set for the address (which is overridden by the transport).
1969 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
1972 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
1973 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
1974 because it is too late at that time, and has no effect.
1976 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
1977 security issue with \' (bugzilla #107). I could not use the
1978 PQescapeStringConn() function, because it needs a PGconn value as one of
1981 PH/08 When testing addresses using -bt, indicate those final addresses that
1982 are duplicates that would not cause an additional delivery. At least one
1983 person was confused, thinking that -bt output corresponded to deliveries.
1984 (Suppressing duplicates isn't a good idea as you lose the information
1985 about possibly different redirections that led to the duplicates.)
1987 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
1988 systems where poll() doesn't work, in particular OS X.
1990 PH/10 Added more information to debugging output for retry time not reached.
1992 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
1993 operations in malware.c.
1995 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
1998 PH/13 If write_rejectlog was set false when logging was sent to syslog with
1999 syslog_duplication set false, log lines that would normally be written
2000 both the the main log and to the reject log were not written to syslog at
2003 PH/14 In the default configuration, change the use of "message" in ACL warn
2004 statements to "add_header".
2006 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2007 not followed by a command (e.g. "seen endif").
2009 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2010 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2013 PH/17 Added extra conditions to the default value of delay_warning_condition
2017 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2018 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2019 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2022 The Auto-Submitted: and various List- headers are standardised, whereas I
2023 don't think Precedence: ever was.
2025 PH/18 Refactored debugging code in route_finduser() to show more information,
2026 in particular, the error code if getpwnam() issues one.
2028 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2029 This is apparently needed in addition to the PH/07 change above to avoid
2030 any possible encoding problems.
2032 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2033 but not after initializing Perl.
2035 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2036 output them only if debugging. By default they are written stderr,
2037 apparently, which is not desirable.
2039 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2042 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2045 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2047 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2048 authenticated or an ident call has been made. Suppress the default
2049 values for $authenticated_id and $authenticated_sender (but permit -oMai
2050 and -oMas) when testing with -bh.
2052 PH/24 Re-jigged the order of the tests in the default configuration so that the
2053 tests for valid domains and recipients precede the DNS black list and CSA
2054 tests, on the grounds that those ones are more expensive.
2056 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2057 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2058 This bug exists in every version of Exim that I still have, right back to
2061 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2062 However, an attempt to turn on case-sensitivity in a regex key by
2063 including (?-i) didn't work because the subject string was already
2064 lowercased, and the effects were non-intuitive. It turns out that a
2065 one-line patch can be used to allow (?-i) to work as expected.
2071 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2072 other effects) broke the use of negated acl sub-conditions.
2074 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2077 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2078 "Deny" causes Exim to reject the incoming connection with a 554 error.
2079 Unfortunately, if there is a major crisis, such as a disk failure,
2080 tcp-wrappers gives "deny", whereas what one would like would be some
2081 kind of temporary error. A kludge has been added to help with this.
2082 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2083 554 error is used if errno is still zero or contains ENOENT (which occurs
2084 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2087 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2089 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2090 errors. Otherwise a message that provokes a temporary error (when other
2091 messages do not) can cause a whole host to time out.
2093 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2094 addresses were routed directly to files or pipes from a redirect router.
2095 File deliveries just didn't batch; pipe deliveries might have suffered
2098 PH/06 A failure to get a lock for a hints database would erroneously always say
2099 "Failed to get write lock", even when it was really a read lock.
2101 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2102 of 0600. This has been changed to use the value of the lockfile_mode
2103 option (which defaults to 0600).
2105 PH/08 Applied small patch from the Sieve maintainer.
2107 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2108 folder from quota calculations, a direct delivery into this folder messed
2109 up the contents of the maildirsize file. This was because the regex was
2110 used only to exclude .Trash (or whatever) when the size of the mailbox
2111 was calculated. There was no check that a delivery was happening into an
2112 excluded directory. This bug has been fixed by ignoring all quota
2113 processing for deliveries into excluded directories.
2115 PH/10 Added the maildirfolder_create_regex option to appendfile.
2121 PH/01 The code for finding all the local interface addresses on a FreeBSD
2122 system running IPv6 was broken. This may well have applied to all BSD
2123 systems, as well as to others that have similar system calls. The broken
2124 code found IPv4 interfaces correctly, but gave incorrect values for the
2125 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2126 that it would not match correctly against @[] and not recognize the IPv6
2129 PH/02 The ipliteral router was not recognizing addresses of the form user@
2130 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2132 PH/03 Added disable_ipv6.
2134 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2135 decoded form, because it is most often used to construct To: headers
2136 lines in autoreplies, and the decoded form may well be syntactically
2137 invalid. However, $reply_address has leading white space removed, and all
2138 newlines turned into spaces so that the autoreply transport does not
2141 PH/05 If group was specified without a user on a router, and no group or user
2142 was specified on a transport, the group from the router was ignored.
2144 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2145 for visible compile-time settings that can be used to change these
2146 numbers, for those that want even more. Backwards compatibility with old
2147 spool files has been maintained. However, going back to a previous Exim
2148 release will lost any variables that are in spool files.
2150 PH/07 Two small changes when running in the test harness: increase delay when
2151 passing a TCP/IP connection to a new process, in case the original
2152 process has to generate a bounce, and remove special handling of
2153 127.0.0.2 (sic), which is no longer necessary.
2155 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2156 be the same on different OS.
2158 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2161 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2162 whether --show-vars was specified or not
2164 JJ/02 exipick: Added support for new ACL variable spool format introduced
2167 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2168 syntactically invalid From: or Reply-to: line, and a filter used this to
2169 generate an autoreply, and therefore failed to obtain an address for the
2170 autoreply, Exim could try to deliver to a non-existent relative file
2171 name, causing unrelated and misleading errors. What now happens is that
2172 it logs this as a hard delivery error, but does not attempt to create a
2175 PH/11 The exinext utility has a -C option for testing purposes, but although
2176 the given file was scanned by exinext itself; it wasn't being passed on
2177 when Exim was called.
2179 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2180 an end-of-file indication when reading a command response.
2182 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2183 compiled. In many other places in Exim, IPv6 addresses are always
2184 recognized, so I have changed this. It also means that IPv4 domain
2185 literals of the form [IPV4:n.n.n.n] are now always recognized.
2187 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2188 used if the router is not running as root, for example, when verifying at
2189 ACL time, or when using -bh. The debugging output from this situation was
2190 non-existent - all you got was a failure to exec. I have made two
2193 (a) Failures to set uid/gid, the current directory, or a process leader
2194 in a subprocess such as that created by queryprogram now generate
2195 suitable debugging ouput when -d is set.
2197 (b) The queryprogram router detects when it is not running as root,
2198 outputs suitable debugging information if -d is set, and then runs
2199 the subprocess without attempting to change uid/gid.
2201 PH/15 Minor change to Makefile for building test_host (undocumented testing
2204 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2205 additional section of a DNS packet that returns MX or SRV records.
2206 Instead, it always explicitly searches for A/AAAA records. This avoids
2207 major problems that occur when a DNS server includes only records of one
2208 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2209 fixed another bug: if SRV records were looked up and the corresponding
2210 address records were *not* found in the additional section, the port
2211 values from the SRV records were lost.
2213 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2214 using the correct key (the original address) when searching the retry
2215 rules in order to find which one to use for generating the retry hint.
2217 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2218 adding the default one. Similarly, if it contains a Reply-To: header, the
2219 errors_reply_to option, if set, is not used.
2221 PH/19 When calculating a retry time, Exim used to measure the "time since
2222 failure" by looking at the "first failed" field in the retry record. Now
2223 it does not use this if it is later than than the arrival time of the
2224 message. Instead it uses the arrival time. This makes for better
2225 behaviour in cases where some deliveries succeed, thus re-setting the
2226 "first failed" field. An example is a quota failure for a huge message
2227 when small messages continue to be delivered. Without this change, the
2228 "time since failure" will always be short, possible causing more frequent
2229 delivery attempts for the huge message than are intended.
2230 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2232 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2233 $1, $2, $3) because the numerical variables can be reset during some
2234 expansion items (e.g. "match"), thereby losing the authentication data.
2236 PH/21 Make -bV show the size of off_t variables so that the test suite can
2237 decide whether to run tests for quotas > 2G.
2239 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2240 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2241 filecount value is greater than 2G or if a quota value is greater than 2G
2242 on a system where the size of off_t is not greater than 4, a panic error
2245 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2246 never match. The debug and -bh output now contains an explicit error
2247 message indicating a malformed IPv4 address or mask.
2249 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2250 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2251 PH/23 above applies.
2253 PH/25 Do not write to syslog when running in the test harness. The only
2254 occasion when this arises is a failure to open the main or panic logs
2255 (for which there is an explicit test).
2257 PH/26 Added the /no_tell option to "control=freeze".
2259 PH/27 If a host name lookup failed very early in a connection, for example, if
2260 the IP address matched host_lookup and the reverse lookup yielded a name
2261 that did not have a forward lookup, an error message of the form "no IP
2262 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2263 could be logged. Now it outputs the IP address instead of "NULL".
2265 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2266 allows the sender and the authenticated sender to be set when
2267 submitting a message from within Exim. Since child_open_exim() is
2268 documented for local_scan(), the new function should be too.
2270 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2271 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2272 results in an empty string is now treated as unset.
2274 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2276 PH/31 Added sender_verify_fail logging option.
2278 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2279 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2280 tidied the source and removed it altogether.
2282 PH/33 When a queue run was abandoned because the load average was too high, a
2283 log line was always written; now it is written only if the queue_run log
2284 selector is set. In addition, the log line for abandonment now contains
2285 information about the queue run such as the pid. This is always present
2286 in "start" and "stop" lines but was omitted from the "abandon" line.
2288 PH/34 Omit spaces between a header name and the colon in the error message that
2289 is given when verify = headers_syntax fails (if there are lots of them,
2290 the message gets confusing).
2292 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2293 names, as there are now some PTR records that contain slashes. This check
2294 is only to protect against broken name servers that fall over on strange
2295 characters, so the fact that it applies to all lookups doesn't matter.
2297 PH/36 Now that the new test suite is complete, we can remove some of the
2298 special code in Exim that was needed for the old test suite. For example,
2299 sorting DNS records because real resolvers return them in an arbitrary
2300 order. The new test suite's fake resolver always returns records in the
2303 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2304 bounces) except when queue_only is set, to avoid logging races between
2305 the different processes.
2307 PH/38 Panic-die if .include specifies a non-absolute path.
2309 PH/39 A tweak to the "H" retry rule from its user.
2311 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2312 a label. They prevented compilation on older perls.
2314 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2315 a warning to be raised on newish perls.
2317 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2318 on queue. Changes to match documented behaviour of showing count of
2319 messages matching specified criteria.
2321 PH/40 Changed the default ident timeout from 30s to 5s.
2323 PH/41 Added support for the use of login_cap features, on those BSD systems
2324 that have them, for controlling the resources used by pipe deliveries.
2326 PH/42 The content-scanning code uses fopen() to create files in which to put
2327 message data. Previously it was not paying any attention to the mode of
2328 the files. Exim runs with umask(0) because the rest of the code creates
2329 files with open(), and sets the required mode explicitly. Thus, these
2330 files were ending up world-writeable. This was not a big issue, because,
2331 being within the spool directory, they were not world-accessible. I have
2332 created a function called modefopen, which takes an additional mode
2333 argument. It sets umask(777), creates the file, chmods it to the required
2334 mode, then resets the umask. All the relevant calls to fopen() in the
2335 content scanning code have been changed to use this function.
2337 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2338 to 24 hours. This avoids potential overflow problems when processing G
2339 and H retry rules. I suspect nobody ever tinkers with this value.
2341 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2343 PH/45 When the plaintext authenticator is running as a client, the server's
2344 challenges are checked to ensure they are valid base64 strings. By
2345 default, the authentication attempt is cancelled if an invalid string is
2346 received. Setting client_ignore_invalid_base64 true ignores these errors.
2347 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2348 they are received. Thus, the responses can be made to depend on the
2349 challenges. If an invalid string is ignored, an empty string is placed in
2352 PH/46 Messages that are created by the autoreply transport now contains a
2353 References: header, in accordance with RFCs 2822 and 3834.
2355 PH/47 Added authenticated_sender_force to the smtp transport.
2357 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2359 PH/49 Installed latest patch from the Sieve maintainer.
2361 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2362 was also set, the appendfile transport was unnecessarily scanning a
2363 directory of message files (e.g. for maildir delivery) to find the count
2364 of files (along with the size), even though it did not need this
2365 information. It now does the scan only if it needs to find either the
2366 size of the count of files.
2368 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2370 PH/52 Two bugs concerned with error handling when the smtp transport is
2373 (i) Exim was not creating retry information for temporary errors given
2374 for individual recipients after the DATA command when the smtp transport
2375 was used in LMTP mode. This meant that they could be retried too
2376 frequently, and not timed out correctly.
2378 (ii) Exim was setting the flag that allows error details to be returned
2379 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2380 recipients that were returned after the DATA command.
2382 PH/53 This is related to PH/52, but is more general: for any failing address,
2383 when detailed error information was permitted to be returned to the
2384 sender, but the error was temporary, then after the final timeout, only
2385 "retry timeout exceeded" was returned. Now it returns the full error as
2386 well as "retry timeout exceeded".
2388 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2389 do this, and (what is worse) MTAs that accept it.
2391 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2392 will now be deprecated.
2394 PH/56 New os.c-cygwin from the Cygwin maintainer.
2396 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2397 formats (previously only available in exim formats via -bpr, -bpru,
2398 and -bpra. Now also available in native and exiqgrep formats)
2400 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2401 with very large, slow to parse queues
2403 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2405 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2407 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2408 responses to authentication challenges, though it was showing the
2409 challenges; (ii) I've removed the CR characters from the debug output for
2412 PH/58 Allow for the insertion of a newline as well as a space when a string
2413 is turned into more than one encoded-word during RFC 2047 encoding. The
2414 Sieve code now uses this.
2416 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2417 data_4xx, lost_connection, tls_required.
2419 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2420 message was being sent as an SMTP response.
2422 PH/61 Add -l and -k options to exicyclog.
2424 PH/62 When verifying, if an address was redirected to one new address, so that
2425 verification continued, and the new address failed or deferred after
2426 having set something in $address_data, the value of $address_data was not
2427 passed back to the ACL. This was different to the case when no
2428 redirection occurred. The value is now passed back in both cases.
2430 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2431 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2432 use login_cap.h, so on its own it isn't the distinguishing feature. The
2433 new name refers directly to the setclassresources() function.
2435 PH/65 Added configuration files for NetBSD3.
2437 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2439 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2440 is preferred over IPv4.
2442 PH/68 The bounce_return_message and bounce_return_body options were not being
2443 honoured for bounces generated during the reception of non-SMTP messages.
2444 In particular, this applied to messages rejected by the ACL. This bug has
2445 been fixed. However, if bounce_return_message is true and bounce_return_
2446 body is false, the headers that are returned for a non-SMTP message
2447 include only those that have been read before the error was detected.
2448 (In the case of an ACL rejection, they have all been read.)
2450 PH/69 The HTML version of the specification is now built in a directory called
2451 spec_html instead of spec.html, because the latter looks like a path with
2452 a MIME-type, and this confuses some software.
2454 PH/70 Catch two compiler warnings in sieve.c.
2456 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2457 function verify_get_ident() calls ip_connect() to connect a socket, but
2458 if the "connect()" function timed out, ip_connect() used to close the
2459 socket. However, verify_get_ident() also closes the socket later, and in
2460 between Exim writes to the log, which may get opened at this point. When
2461 the socket was closed in ip_connect(), the log could get the same file
2462 descriptor number as the socket. This naturally causes chaos. The fix is
2463 not to close the socket in ip_connect(); the socket should be closed by
2464 the function that creates it. There was only one place in the code where
2465 this was missing, in the iplookup router, which I don't think anybody now
2466 uses, but I've fixed it anyway.
2468 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2469 well as to direct DNS lookups. Otherwise the handling of names in host
2470 lists is inconsistent and therefore confusing.
2476 PH/01 Two changes to the default runtime configuration:
2478 (1) Move the checks for relay_from_hosts and authenticated clients from
2479 after to before the (commented out) DNS black list checks.
2481 (2) Add control=submission to the relay_from_hosts and authenticated
2482 clients checks, on the grounds that messages accepted by these
2483 statements are most likely to be submissions.
2485 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2487 (1) Generate an error if the third argument for the ${prvs expansion is
2490 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2493 (3) Reset the variables that are obtained from the first argument of
2494 ${prvscheck and used in the second argument before leaving the code,
2495 because their memory is reclaimed, so using them afterwards may do
2498 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2499 one (it's much easier than Tom thought :-).
2501 (5) Because of (4), we can now allow for the use of $prvscheck_result
2502 inside the third argument.
2504 PH/03 For some reason, the default setting of PATH when running a command from
2505 a pipe transport was just "/usr/bin". I have changed it to
2508 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2509 anything to be listed in the output from -bV.
2511 PH/05 When a filter generated an autoreply, the entire To: header line was
2512 quoted in the delivery log line, like this:
2514 => >A.N.Other <ano@some.domain> <original@ddress> ...
2516 This has been changed so that it extracts the operative address. There
2517 may be more than one such address. If so, they are comma-separated, like
2520 => >ano@some.domain,ona@other.domain <original@ddress> ...
2522 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2523 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2524 not being looked up in the rDNS to get a host name, Exim was showing the
2525 IP address twice in Received: lines, even though the IP addresses were
2526 identical. For example:
2528 Received: from [1.2.3.4] (helo=[1.2.3.4])
2530 However, if the real host name was known, it was omitting the HELO data
2531 if it matched the actual IP address. This has been tidied up so that it
2532 doesn't show the same IP address twice.
2534 PH/07 When both +timestamp and +memory debugging was on, the value given by
2535 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2536 called by the debug printing, thereby overwriting the timestamp buffer.
2537 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2539 PH/08 When the original message was included in an autoreply transport, it
2540 always said "this is a copy of the message, including all the headers",
2541 even if body_only or headers_only was set. It now gives an appropriate
2544 PH/09 Applied a patch from the Sieve maintainer which:
2546 o fixes some comments
2547 o adds the (disabled) notify extension core
2548 o adds some debug output for the result of if/elsif tests
2549 o points to the current vacation draft in the documentation
2550 and documents the missing references header update
2554 o fixes a bug in processing the envelope test (when testing
2555 multiple envelope elements, the last element determinted the
2558 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2559 Electronic Mail") by including:
2561 Auto-submitted: auto-generated
2563 in the messages that it generates (bounce messages and others, such as
2564 warnings). In the case of bounce messages for non-SMTP mesages, there was
2565 also a typo: it was using "Auto_submitted" (underscore instead of
2566 hyphen). Since every message generated by Exim is necessarily in response
2567 to another message, thes have all been changed to:
2569 Auto-Submitted: auto-replied
2571 in accordance with these statements in the RFC:
2573 The auto-replied keyword:
2575 - SHOULD be used on messages sent in direct response to another
2576 message by an automatic process,
2578 - MUST NOT be used on manually-generated messages,
2580 - MAY be used on Delivery Status Notifications (DSNs) and Message
2581 Disposition Notifications (MDNs),
2583 - MUST NOT be used on messages generated by automatic or periodic
2584 processes, except for messages which are automatic responses to
2587 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2588 to the default Received: header definition.
2590 PH/12 Added log selector acl_warn_skipped (default on).
2592 PH/13 After a successful wildlsearch lookup, discard the values of numeric
2593 variables because (a) they are in the wrong storage pool and (b) even if
2594 they were copied, it wouldn't work properly because of the caching.
2596 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2597 checking when decoding. Apparently there are clients that generate
2598 overlong encoded strings. Why am I not surprised?
2600 PH/15 If the first argument of "${if match_address" was not empty, but did not
2601 contain an "@" character, Exim crashed. Now it writes a panic log message
2602 and treats the condition as false.
2604 PH/16 In autoreply, treat an empty string for "once" the same as unset.
2606 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
2607 extension "envelope-auth". The code is finished and in agreement with
2608 other implementations, but there is no documentation so far and in fact,
2609 nobody wrote the draft yet. This extension is currently #undef'ed, thus
2610 not changing the active code.
2612 Print executed "if" and "elsif" statements when debugging is used. This
2613 helps a great deal to understand what a filter does.
2615 Document more things not specified clearly in RFC3028. I had all this
2616 sorted out, when out of a sudden new issues came to my mind. Oops."
2618 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
2621 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
2622 canonical form (as documented). However, after a host name lookup from
2623 the IP address, check_host() was doing a simple string comparison with
2624 addresses acquired from the DNS when checking that the found name did
2625 have the original IP as one of its addresses. Since any found IPv6
2626 addresses are likely to be in abbreviated form, the comparison could
2627 fail. Luckily, there already exists a function for doing the comparison
2628 by converting both addresses to binary, so now that is used instead of
2629 the text comparison.
2631 PH/20 There was another similar case to PH/19, when a complete host name was
2632 given in a host list; looking up its IP address could give an abbreviated
2633 form, whereas the current host's name might or might not be abbreviated.
2634 The same fix has been applied.
2640 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
2641 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
2644 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
2645 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
2647 PH/03 Typo: missing ".o" in src/pcre/Makefile.
2649 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
2650 header line, restrict the check to what is listed in RFCs 2369 and 2929.
2651 Also, for "Auto-Submitted", treat anything other than "no" as
2652 non-personal, in accordance with RFC 3834. (Previously it treated
2653 anything starting "auto-" as non-personal.)
2655 TF/01 The control=submission/name=... option had a problem with syntax
2656 errors if the name included a slash character. The /name= option
2657 now slurps the rest of the string, so it can include any characters
2658 but it must come last in the list of options (after /sender_retain
2661 PH/05 Some modifications to the interface to the fake nameserver for the new
2669 TK/01 Added the "success_on_redirect" address verification option. See
2670 NewStuff for rationale and an example.
2672 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
2674 PH/02 Patch to exigrep to allow it to work on syslog lines.
2676 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
2677 fread() to skip over the body file's header line, because in Cygwin the
2678 header line is locked and is inaccessible.
2680 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
2681 co-exist for some time) to make it clear that it is the Exim ID that is
2682 referenced, not the Message-ID: header line.
2684 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
2685 string_format() function, because snprintf() does not exist on all
2688 PH/06 The use of forbid_filter_existstest now also locks out the use of the
2689 ${stat: expansion item.
2691 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
2692 protocol synchronization error", to keep the pedants happy.
2694 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
2695 well as for IRIX systems, when gcc is being used. See the host.c source
2698 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
2700 PH/10 Named domain lists were not working if used in a queue_smtp_domains
2703 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
2704 transport and to the smtp transport in LMTP mode.
2706 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
2708 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
2709 run a filter in a subprocess. This could lead to confusion in subsequent
2710 lookups in the parent process. There should also be a search_tidyup() at
2711 the end of the subprocess.
2713 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
2714 only if the host matched helo_try_verify_hosts, which caused the
2715 verification to occur when the EHLO/HELO command was issued. The ACL just
2716 tested the remembered result. Now, if a previous verification attempt has
2717 not happened, "verify = helo" does it there and then.
2719 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
2721 TK/03 Fix log output including CR from clamd.
2723 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
2724 exist provoked a memory error which could cause a segfault.
2726 PH/15 Installed PCRE 6.2
2728 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
2730 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
2731 of the problem. Specifically, suggested +O2 rather than +O1 for the
2734 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
2736 PH/20 If a delivery was routed to a non-standard port by means of an SRV
2737 record, the port was not correctly logged when the outgoing_port log
2738 selector was set (it logged the transort's default port).
2740 PH/21 Added support for host-specific ports to manualroute, queryprogram,
2741 fallback_hosts, and "hosts" in the smtp transport.
2743 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
2744 host errors such as "Connection refused".
2746 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
2747 authentication with radiusclient 0.4.9:
2749 - Error returned from rc_read_config was caught wrongly
2750 - Username/password not passed on to radius server due to wrong length.
2752 The presumption is that some radiusclient API changes for 4.51/PH/17
2753 were not taken care of correctly. The code is still untested by me (my
2754 Linux distribution still has 0.3.2 of radiusclient), but it was
2755 contributed by a Radius user.
2757 PH/24 When doing a callout, the value of $domain wasn't set correctly when
2758 expanding the "port" option of the smtp transport.
2760 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
2761 while reading a MIME header. Thanks to Tom Hughes for a patch.
2763 PH/24 Include config.h inside local_scan.h so that configuration settings are
2766 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
2767 that in an ACL for RCPT (for example), you can examine exactly what was
2770 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
2771 commands, but it was not correctly comparing the address with the actual
2772 client host address. Thus, it would show the EHLO address in Received:
2773 header lines when this was not necessary.
2775 PH/27 Added the % operator to ${eval:}.
2777 PH/28 Exim tries to create and chdir to its spool directory when it starts;
2778 it should be ignoring failures (because with -C, for example, it has lost
2779 privilege). It wasn't ignoring creation failures other than "already
2782 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
2785 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
2786 because an input line was too long, either on its own, or by virtue of
2787 too many continuations, the temporary file was not being removed, and the
2788 return code was incorrect.
2790 PH/31 Missing "BOOL" in function definition in filtertest.c.
2792 PH/32 Applied Sieve patches from the maintainer.
2794 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
2796 PH/33 Added "verify = not_blind".
2798 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
2799 Local/Makefile (with some defaults set). These are used in built scripts
2800 such as exicyclog, but they have never been used in the exim_install
2801 script (though there are many overriding facilities there). I have
2802 arranged that the exim_install script now takes note of these two
2805 PH/35 Installed configuration files for Dragonfly.
2807 PH/36 When a locally submitted message by a trusted user did not contain a
2808 From: header, and the sender address was obtained from -f or from an SMTP
2809 MAIL command, and the trusted user did not use -F to supply a sender
2810 name, $originator_name was incorrectly used when constructing a From:
2811 header. Furthermore, $originator_name was used for submission mode
2812 messages from external hosts without From: headers in a similar way,
2813 which is clearly wrong.
2815 PH/37 Added control=suppress_local_fixups.
2817 PH/38 When log_selector = +received_sender was set, and the addition of the
2818 sender made the log line's construction buffer exactly full, or one byte
2819 less than full, an overflow happened when the terminating "\n" was
2822 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
2823 when the result of a list match is failure because a DNS lookup failed.
2825 PH/40 RM_COMMAND is now used in the building process.
2827 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
2828 the "build-* directories that it finds.
2830 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
2831 address in a domain literal was a prefix of an interface address.
2833 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
2834 when verifying a sender address, unless rewrite_headers is false.
2836 PH/44 Wrote a long comment about why errors_to addresses are verified as
2837 recipients, not senders.
2839 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
2840 the ratelimit ACL was added.
2842 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
2844 PH/46 Added extra information about PostgreSQL errors to the error string.
2846 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
2847 avoiding the need to install special zones in a real server. This is
2848 backwards compatible; if it can't find the fake resolver, it drops back.
2849 Thus, both old and new test suites can be run.
2851 TF/02 Added util/ratelimit.pl
2853 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
2854 clock is set back in time.
2856 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
2857 Candler <B.Candler@pobox.com>.
2859 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
2860 OK for addresses that are the result of redirections.
2862 PH/48 A number of further additions for the benefit of the new test suite,
2863 including a fake gethostbyname() that interfaces to the fake DNS resolver
2866 TF/06 The fix for widen_domains has also been applied to qualify_single and
2867 search_parents which are the other dnslookup options that can cause
2870 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
2873 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
2875 TF/07 Exim produced the error message "an SRV record indicated no SMTP
2876 service" if it encountered an MX record with an empty target hostname.
2877 The message is now "an MX or SRV record indicated no SMTP service".
2879 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
2880 if the DNS of the sending site is misconfigured. This is quite a
2881 common situation. This change restores the behaviour of treating a
2882 helo verification defer as a failure.
2884 PH/51 If self=fail was set on a router, the bounce message did not include the
2885 actual error message.
2891 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
2893 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
2894 command itself ended in error, the underlying message about the transport
2895 filter timeout was being overwritten with the pipe command error. Now the
2896 underlying error message should be appended to the second error message.
2898 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
2900 PH/02 Reduce the amount of output that "make" produces by default. Full output
2901 can still be requested.
2903 PH/03 The warning log line about a condition test deferring for a "warn" verb
2904 was being output only once per connection, rather than after each
2905 occurrence (because it was using the same function as for successful
2906 "warn" verbs). This seems wrong, so I have changed it.
2908 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
2909 it should not have, which might have caused a crash in the right
2910 circumstances, but probably never did.
2912 PH/04 Installed a modified version of Tony Finch's patch to make submission
2913 mode fix the return path as well as the Sender: header line, and to
2914 add a /name= option so that you can make the user's friendly name appear
2917 TF/03 Added the control = fakedefer ACL modifier.
2919 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
2920 Mark Lowes for thorough testing.
2922 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
2924 TK/03 Merged latest SRS patch from Miles Wilton.
2926 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
2927 with the definition in sysexits.h (which is #included earlier).
2928 Fortunately, Exim does not actually use EX_OK. The code used to try to
2929 preserve the sysexits.h value, by assumimg that macro definitions were
2930 scanned for macro replacements. I have been disabused of this notion,
2931 so now the code just undefines EX_OK before #including unistd.h.
2933 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
2934 in the smtp transport. When a block could not be written in a single
2935 write() function, the timeout was being re-applied to each part-write.
2936 This seems wrong - if the receiver was accepting one byte at a time it
2937 would take for ever. The timeout is now adjusted when this happens. It
2938 doesn't have to be particularly precise.
2940 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
2941 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
2943 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
2944 without a domain if the check to <postmaster@domain> fails.
2946 SC/01 Eximstats: added -xls and the ability to specify output files
2947 (patch written by Frank Heydlauf).
2949 SC/02 Eximstats: use FileHandles for outputing results.
2951 SC/03 Eximstats: allow any combination of xls, txt, and html output.
2953 SC/04 Eximstats: fixed display of large numbers with -nvr option
2955 SC/05 Eximstats: fixed merging of reports with empty tables.
2957 SC/06 Eximstats: added the -include_original_destination flag
2959 SC/07 Eximstats: removed tabs and trailing whitespace.
2961 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
2963 TK/06 MBOX spool code: Add real "From " MBOX separator line
2964 so the .eml file is really in mbox format (even though
2965 most programs do not really care). Patch from Alex Miller.
2967 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
2968 The latter is generated from $received_to and is only set if the
2969 message has one envelope recipient. SA can use these headers,
2970 obviously out-of-the-box. Patch from Alex Miller.
2972 PH/08 The ${def test on a variable was returning false if the variable's
2973 value was "0", contrary to what the specification has always said!
2974 The result should be true unless the variable is empty.
2976 PH/09 The syntax error of a character other than { following "${if
2977 def:variable_name" (after optional whitespace) was not being diagnosed.
2978 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
2979 accidental colon was present, for example, could give incorrect results.
2981 PH/10 Tidied the code in a number of places where the st_size field of a stat()
2982 result is used (not including appendfile, where other changes are about
2985 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
2986 This involved changing a lot of size variables from int to off_t. It
2987 should work with maildirs and everything.
2989 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
2990 spamd dying while we are connected to it.
2992 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
2995 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
2996 function for BDB 4.3.
2998 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3000 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3001 the sources for PCRE have been re-arranged and more files are now
3004 PH/15 The code I had for printing potentially long long variables in PH/11
3005 above was not the best (it lost precision). The length of off_t variables
3006 is now inspected at build time, and an appropriate printing format (%ld
3007 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3008 to be "long long int" or "long int". This is needed for the internal
3009 formatting function string_vformat().
3011 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3012 the configuration file to be ":syslog", then the script "guesses" where
3013 the logs files are, rather than using the compiled in default. In our
3014 case the guess is not the same as the compiled default, so the script
3015 suddenly stopped working when I started to use syslog. The patch checks
3016 to see if log_file_path is "". If so, it attempts to read it from exim
3017 with no configuration file to get the compiled in version, before it
3018 falls back to the previous guessing code."
3020 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3021 implementing BATV in an Exim configuration. See NewStuff for the gory
3024 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3025 Makefile that are specific to HP-UX.
3027 PH/18 If the "use_postmaster" option was set for a recipient callout together
3028 with the "random" option, the postmaster address was used as the MAIL
3029 FROM address for the random test, but not for the subsequent recipient
3030 test. It is now used for both.
3032 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3033 patch removes a few documentation additions to RFC 3028, because the
3034 latest draft now contains them. It adds the new en;ascii-case comparator
3035 and a new error check for 8bit text in MIME parts. Comparator and
3036 require names are now matched exactly. I enabled the subaddress
3037 extension, but it is not well tested yet (read: it works for me)."
3039 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3040 rework some of the code of TK/09 above to avoid the hardwired use of
3041 "%lld" and "long long". Replaced the call to snprintf() with a call to
3044 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3045 records point to non-existent hosts", "retry timeout exceeded", and
3046 "retry time not reached for any host after a long failure period".
3048 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3049 experimental DomainKeys support:
3051 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3052 (2) On an error such as an illegally used "control", the wrong name for
3053 the control was given.
3055 These problems did NOT occur unless DomainKeys support was compiled.
3057 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3059 PH/24 Added ${if match_ip condition.
3061 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3062 databases so that it will be absolutely obvious if a crash occurs in the
3063 DB library. This is a regular occurrence (often caused by mis-matched
3066 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3067 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3068 picked up on a user's system that detects such things. There doesn't seem
3069 to be a gcc warning option for this - only an attribute that has to be
3070 put on the function's prototype. It seems that in Fedora Core 4 they have
3071 set this on a number of new functions. No doubt there will be more in due
3074 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3075 specify a transport. However, if an address that was verified by such a
3076 router was the subject of a callout, Exim crashed because it tried to
3077 read the rcpt_include_affixes from the non-existent transport. Now it
3078 just assumes that the setting of that option is false. This bug was
3079 introduced by 4.51/PH/31.
3081 PH/28 Changed -d+all to exclude +memory, because that information is very
3082 rarely of interest, but it makes the output a lot bigger. People tend to
3083 do -d+all out of habit.
3085 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3086 code in os-type was giving problems when libc.so lives in lib64, like on
3089 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3090 aren't the modern standard, and it seems that some systems' include files
3091 don't always have them. Exim was already checking for some of the newer
3092 ones like T_AAAA, and defining it itself. I've added checks for all the
3093 record types that Exim uses.
3095 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3096 not automatically generating a new one, as it is supposed to. This
3097 prevented TLS from working. If the file did exist, but contained invalid
3098 data, a new version was generated, as expected. It was only the case of a
3099 non-existent file that was broken.
3101 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3102 with a change in libdomainkeys > 0.64.
3104 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3105 from DNS. If the selector record carries the flag, it now has
3106 precedence over the domain-wide flag.
3108 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3110 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3111 the use of a port name that isn't defined in /etc/services) occurred, the
3112 message was deferred as in a normal delivery, and thus remained on the
3113 spool, instead of being failed because of the mua_wrapper setting. This
3114 is now fixed, and I tidied up some of the mua_wrapper messages at the
3117 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3118 the messages in a hash of arrays rather than using individual hashes.
3119 This is a bit cleaner and results in dramatic memory savings, albeit
3120 at a slight CPU cost.
3122 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3123 as requested by Marc Sherman.
3125 SC/10 Eximstats: added histograms for user specified patterns as requested
3128 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3130 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3131 fopen() in the content-scanning modules that did not already have it.
3137 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3138 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3140 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3142 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3144 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3145 to test Sieve filters that use "vacation".
3147 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3148 that changes the way the GnuTLS parameters are stored in the cache file.
3149 The new format can be generated externally. For backward compatibility,
3150 if the data in the cache doesn't make sense, Exim assumes it has read an
3151 old-format file, and it generates new data and writes a new file. This
3152 means that you can't go back to an older release without removing the
3155 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3156 work if there are any delivery delays because "one_time" forces the
3157 parent to be marked "delivered", so its unseen clone is never tried
3158 again. For this reason, Exim now forbids the simultaneous setting of
3161 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3162 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3163 entry if you want to know the details. The fix, however, neglected to
3164 consider the case when local delivery batching is involved. The test for
3165 "previously delivered" was not happening when checking to see if an
3166 address could be batched with a previous (undelivered) one; under
3167 certain circumstances this could lead to multiple deliveries to the same
3170 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3171 in its include files, and this causes problems building Exim.
3173 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3174 header_syntax) but Exim was just ignoring anything given after a slash.
3175 In particular, this caused confusion with an attempt to use "verify =
3176 reverse_host_lookup/defer_ok". An error is now given when options are
3177 supplied for verify items that do not have them. (Maybe reverse_host_
3178 lookup should have a defer_ok option, but that's a different point.)
3180 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3181 defined by RFC 821) to 2048, because there were problems with some AUTH
3182 commands, and RFC 1869 says the size should be increased for extended
3183 SMTP commands that take arguments.
3185 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3188 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3189 "unknown" error; now it says that the functionality isn't in the binary.
3191 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3192 an address' error message when a string expansion fails (syntax or
3193 whatever). Otherwise the password may appear in the log. Following change
3194 PH/42 below, there is no longer a chance of it appearing in a bounce
3197 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3199 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3200 of its addresses was ever tried. (Bugzilla bug #2.)
3202 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3203 the result incorrectly in the debug output. (It correctly added a newline
3204 to what was transported.)
3206 TF/01 Added $received_time.
3208 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3209 commented out examples of how to interface to a virus scanner and to
3210 SpamAssassin. Also added commented examples of av_scanner and
3211 spamd_address settings.
3213 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3214 and controls are allowed in which ACLs. There were a couple of minor
3215 errors. Some of the entries in the conditions table (which is a table of
3216 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3217 negation of where the condition IS allowed.
3219 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3221 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3222 header file does not have a version number, so I've had to invent a new
3223 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3224 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3225 radiusclient), but it was contributed by a Radius user.
3227 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3228 files or directories, for OpenSSL.
3230 PH/19 When an Exim process that is running as root has to create an Exim log
3231 file, it does so in a subprocess that runs as exim:exim so as to get the
3232 ownership right at creation (otherwise, other Exim processes might see
3233 the file with the wrong ownership). There was no test for failure of this
3234 fork() call, which would lead to the process getting stuck as it waited
3235 for a non-existent subprocess. Forks do occasionally fail when resources
3236 run out. I reviewed all the other calls to fork(); they all seem to check
3239 PH/20 When checking for unexpected SMTP input at connect time (before writing
3240 the banner), Exim was not dealing correctly with a non-positive return
3241 from the read() function. If the client had disconnected by this time,
3242 the result was a log entry for a synchronization error with an empty
3243 string after "input=" when read() returned zero. If read() returned -1
3244 (an event I could not check), uninitialized data bytes were printed.
3245 There were reports of junk text (parts of files, etc) appearing after
3248 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3250 PH/22 Added support for macro redefinition, and (re)definition in between
3251 driver and ACL definitions.
3253 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3254 forgetting to use the resulting value; it was using the unexpanded value.
3256 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3257 hadn't been configured. The fix is from Juergen Kreileder, who
3258 understands it better than I do:
3260 "Here's what I see happening with three configured cyrus_sasl
3261 authenticators configured (plain, login, cram-md5):
3263 On startup auth_cyrus_sasl_init() gets called for each of these.
3264 This means three calls to sasl_listmech() without a specified mech_list.
3265 => SASL tests which mechs of all available mechs actually work
3266 => three warnings about OTP not working
3267 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3269 With the patch, sasl_listmech() also gets called three times. But now
3270 SASL's mech_list option is set to the server_mech specified in the the
3271 authenticator. Or in other words, the answer from sasl_listmech()
3272 gets limited to just the mech you're testing for (which is different
3274 => the return list contains just 'plain' or 'login', 'cram-md5' or
3275 nothing depending on the value of ob->server_mech.
3277 I've just tested the patch: Authentication still works fine,
3278 unavailable mechs specified in the exim configuration are still
3279 caught, and the auth.log warnings about OTP are gone."
3281 PH/25 When debugging is enabled, the contents of the command line are added
3282 to the debugging output, even when log_selector=+arguments is not
3285 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3286 answer is "GNU", and only if the return is "GNU/something" is the answer
3289 PH/27 $acl_verify_message is now set immediately after the failure of a
3290 verification in an ACL, and so is available in subsequent modifiers. In
3291 particular, the message can be preserved by coding like this:
3293 warn !verify = sender
3294 set acl_m0 = $acl_verify_message
3296 Previously, $acl_verify_message was set only while expanding "message"
3297 and "log_message" when a very denied access.
3299 PH/28 Modified OS/os.c-Linux with
3301 -#ifndef OS_LOAD_AVERAGE
3302 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3304 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3305 nomenclature these days.)
3307 PH/29 Installed patch from the Sieve maintainer that adds the options
3308 sieve_useraddress and sieve_subaddress to the redirect router.
3310 PH/30 In these circumstances:
3311 . Two addresses routed to the same list of hosts;
3312 . First host does not offer TLS;
3313 . First host accepts first address;
3314 . First host gives temporary error to second address;
3315 . Second host offers TLS and a TLS session is established;
3316 . Second host accepts second address.
3317 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3318 and peerdn, if requested) that were in fact used only for the second
3321 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3322 attention to any local part prefix or suffix that was matched by the
3323 router that accepted the address. It now behaves in the same way as it
3324 does for delivery: the affixes are removed from the local part unless
3325 rcpt_include_affixes is set on the transport.
3327 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3328 timeout during the DATA phase of an incoming message.
3330 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3331 applied a patch sanctioned by the Sieve maintainer.
3333 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3334 the uid or gid is negative. A case of a negative gid caused this to be
3335 noticed. The fix allows for either to be negative.
3337 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3338 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3341 PH/36 The change PH/12 above was broken. Fixed it.
3343 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3344 the grounds that routing the same address twice would always produce the
3345 same answer. This might have been true once, but it is certainly no
3346 longer true now. Routing a child address may depend on the previous
3347 routing that produced that child. Some complicated redirection strategies
3348 went wrong when messages had multiple recipients, and made Exim's
3349 behaviour dependent on the order in which the addresses were given.
3351 I have moved the duplicate checking until after the routing is complete.
3352 Exim scans the addresses that are assigned to local and remote
3353 transports, and removes any duplicates. This means that more work will be
3354 done, as duplicates will always all be routed, but duplicates are
3355 presumably rare, so I don't expect this is of any significance.
3357 For deliveries to pipes, files, and autoreplies, the duplicate checking
3358 still happens during the routing process, since they are not going to be
3361 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3362 It corrects a timeout issue with spamd. This is Ian's comment: "The
3363 background is that sometimes spamd either never reads data from a
3364 connection it has accepted, or it never writes response data. The exiscan
3365 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3366 blindly assumes that writes won't block so it may never time out."
3368 PH/39 Allow G after quota size as well as K and M.
3370 PH/40 The value set for $authenticated_id in an authenticator may not contain
3371 binary zeroes or newlines because the value is written to log lines and
3372 to spool files. There was no check on this. Now the value is run through
3373 the string_printing() function so that such characters are converted to
3374 printable escape sequences.
3376 PH/41 $message_linecount is a new variable that contains the total number of
3377 lines in the message. Compare $body_linecount, which is the count for the
3380 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3381 bounce and delay warning messages, except in certain special cases, which
3384 (a) An SMTP error message from a remote host;
3385 (b) A message specified in a :fail: redirection;
3386 (c) A message specified in a "fail" command in a system filter;
3387 (d) A message specified in a FAIL return from the queryprogram router;
3388 (e) A message specified by the cannot_route_message router option.
3390 In these cases only, Exim does include the error details in bounce and
3391 warning messages. There are also a few cases where bland messages such
3392 as "unrouteable address" or "local delivery error" are given.
3394 PH/43 $value is now also set for the "else" part of a ${run expansion.
3396 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3397 being worked on, but at least Exim now implements the latest version to
3400 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3401 process to complete was treated as a delivery failure, a timeout while
3402 writing the message to the pipe was logged, but erroneously treated as a
3403 successful delivery. Such timeouts include transport filter timeouts. For
3404 consistency with the overall process timeout, these timeouts are now
3405 treated as errors, giving rise to delivery failures by default. However,
3406 there is now a new Boolean option for the pipe transport called
3407 timeout_defer, which, if set TRUE, converts the failures into defers for
3408 both kinds of timeout. A transport filter timeout is now identified in
3411 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3412 systems where "make" and "gmake" are different, calling "gmake" at top
3413 level broke things. I've arranged for the value of $(MAKE) to be passed
3414 from the Makefile to this script so that it can call the same version of
3418 A note about Exim versions 4.44 and 4.50
3419 ----------------------------------------
3421 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3422 changes of various kinds. As a consequence, a big documentation update was
3423 needed. This delayed the release for rather longer than seemed good, especially
3424 in the light of a couple of (minor) security issues. Therefore, the changes
3425 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3426 release. So 4.44 and 4.50 are in effect two different branches that both start
3429 I have left the 4.50 change log unchanged; it contains all the changes since
3430 4.43. The change log for 4.44 is below; many of its items are identical to
3431 those for 4.50. This seems to be the most sensible way to preserve the
3432 historical information.
3438 1. Minor wording change to the doc/README.SIEVE file.
3440 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3441 computation of the current number of files was incorrect.
3443 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3444 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3447 4. Give more explanation in the error message when the command for a transport
3448 filter fails to execute.
3450 5. There are several places where Exim runs a non-Exim command in a
3451 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3452 was being done only for the command run by the queryprogram router. It is
3453 now done for all such subprocesses. The other cases are: ${run, transport
3454 filters, and the commands run by the lmtp and pipe transports.
3456 6. Added CONFIGURE_GROUP build-time option.
3458 7. Some older OS have a limit of 256 on the maximum number of file
3459 descriptors. Exim was using setrlimit() to set 1000 as a large value
3460 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3461 systems. I've change it so that if it can't get 1000, it tries for 256.
3463 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3464 was an oversight, and furthermore, ever since the addition of extra
3465 controls (e.g. 4.43/32), the checks on when to allow different forms of
3466 "control" were broken. There should now be diagnostics for all cases when a
3467 control that does not make sense is encountered.
3469 9. Added the /retain_sender option to "control=submission".
3471 10. $recipients is now available in the predata ACL (oversight).
3473 11. Tidy the search cache before the fork to do a delivery from a message
3474 received from the command line. Otherwise the child will trigger a lookup
3475 failure and thereby defer the delivery if it tries to use (for example) a
3476 cached ldap connection that the parent has called unbind on.
3478 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3479 of $address_data from the recipient verification was clobbered by the
3480 sender verification.
3482 13. The value of address_data from a sender verification is now available in
3483 $sender_address_data in subsequent conditions in the ACL statement.
3485 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3487 15. Added a new option "connect=<time>" to callout options, to set a different
3490 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3491 was its contents. (It was OK if the option was not defined at all.)
3493 17. A "Completed" log line is now written for messages that are removed from
3494 the spool by the -Mrm option.
3496 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3497 information about exactly what failed.
3499 19. Added -dd to debug only the daemon process.
3501 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3502 handles timeouts, both on the server side and network timeouts. Renamed the
3503 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3505 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3506 It is now set to "smtps".
3508 22. $host_address is now set to the target address during the checking of
3509 ignore_target_hosts.
3511 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3512 being passed; this would have caused $sender_host_name to have been used if
3513 matching the list had actually called for a host name (not very likely,
3514 since this list is usually IP addresses). A host name is now passed as
3517 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3518 code that specifies a non-restarting handler (typically sigaction() in
3519 modern systems) in an attempt to fix a rare and obscure crash bug.
3521 25. Narrowed the window for a race in the daemon that could cause it to ignore
3522 SIGCHLD signals. This is not a major problem, because they are used only to
3523 wake it up if nothing else does.
3525 26. A malformed maildirsize file could cause Exim to calculate negative values
3526 for the mailbox size or file count. Odd effects could occur as a result.
3527 The maildirsize information is now recalculated if the size or filecount
3530 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3531 support for a long time. Removed HAVE_SYS_VFS_H.
3533 28. Installed the latest version of exipick from John Jetmore.
3535 29. In an address list, if the pattern was not a regular expression, an empty
3536 subject address (from a bounce message) matched only if the pattern was an
3537 empty string. Non-empty patterns were not even tested. This was the wrong
3538 because it is perfectly reasonable to use an empty address as part of a
3539 database query. An empty address is now tested by patterns that are
3540 lookups. However, all the other forms of pattern expect the subject to
3541 contain a local part and a domain, and therefore, for them, an empty
3542 address still always fails if the pattern is not itself empty.
3544 30. Exim went into a mad DNS loop when attempting to do a callout where the
3545 host was specified on an smtp transport, and looking it up yielded more
3546 than one IP address.
3548 31. Re-factored the code for checking spool and log partition space into a
3549 function that finds that data and another that does the check. The former
3550 is then used to implement four new variables: $spool_space, $log_space,
3551 $spool_inodes, and $log_inodes.
3553 32. The RFC2047 encoding function was originally intended for short strings
3554 such as real names; it was not keeping to the 75-character limit for
3555 encoded words that the RFC imposes. It now respects the limit, and
3556 generates multiple encoded words if necessary. To be on the safe side, I
3557 have increased the buffer size for the ${rfc2047: expansion operator from
3560 33. It is now permitted to omit both strings after an "if" condition; if the
3561 condition is true, the result is "true". As before, when the second string
3562 is omitted, a false condition yields an empty string. This makes it less
3563 cumbersome to write custom ACL and router conditions.
3565 34. Failure to deliver a bounce message always caused it to be frozen, even if
3566 there was an errors_to setting on the router. The errors_to setting is now
3569 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3570 canonical form (fully expanded) before being placed in
3571 $sender_host_address.
3573 36. The table in the code that translates DNS record types into text (T_A to
3574 "A" for instance) was missing entries for NS and CNAME. It is just possible
3575 that this could have caused confusion if both these types were looked up
3576 for the same domain, because the text type is used as part of Exim's
3577 per-process caching. But the chance of anyone hitting this buglet seems
3580 37. The dnsdb lookup has been extended in a number of ways.
3582 (1) There is a new type, "zns", which walks up the domain tree until it
3583 finds some nameserver records. It should be used with care.
3585 (2) There is a new type, "mxh", which is like "mx" except that it returns
3586 just the host names, not the priorities.
3588 (3) It is now possible to give a list of domains (or IP addresses) to be
3589 looked up. The behaviour when one of the lookups defers can be
3590 controlled by a keyword.
3592 (4) It is now possible to specify the separator character for use when
3593 multiple records are returned.
3595 38. The dnslists ACL condition has been extended: it is now possible to supply
3596 a list of IP addresses and/or domains to be looked up in a particular DNS
3599 39. Added log_selector=+queue_time_overall.
3601 40. When running the queue in the test harness, wait just a tad after forking a
3602 delivery process, to get repeatability of debugging output.
3604 41. Include certificate and key file names in error message when GnuTLS fails
3605 to set them up, because the GnuTLS error message doesn't include the name
3606 of the failing file when there is a problem reading it.
3608 42. Allow both -bf and -bF in the same test run.
3610 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
3612 44. The "Exiscan patch" is now merged into the mainline Exim source.
3614 45. Sometimes the final signoff response after QUIT could fail to get
3615 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3616 < 0 before doing a fflush(). This bug looks as though it goes back to the
3617 introduction of TLS in release 3.20, but "sometimes" must have been rare
3618 because the tests only now provoked it.
3620 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
3621 (this can affect the format of dates).
3623 47. exim_tidydb, when checking for the continued existence of a message for
3624 which it has found a message-specific retry record, was not finding
3625 messages that were in split spool directories. Consequently, it was
3626 deleting retry records that should have stayed in existence.
3628 48. Steve fixed some bugs in eximstats.
3630 49. The SPA authentication driver was not abandoning authentication and moving
3631 on to the next authenticator when an expansion was forced to fail,
3632 contradicting the general specification for all authenticators. Instead it
3633 was generating a temporary error. It now behaves as specified.
3635 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
3636 (the order specifies the preference for clients). The order is now AES256,
3637 AES128, 3DES, ARCFOUR128.
3639 51. Small patch to Sieve code - explicitly set From: when generating an
3642 52. Exim crashed if a remote delivery caused a very long error message to be
3643 recorded - for instance if somebody sent an entire SpamAssassin report back
3644 as a large number of 550 error lines. This bug was coincidentally fixed by
3645 increasing the size of one of Exim's internal buffers (big_buffer) that
3646 happened as part of the Exiscan merge. However, to be on the safe side, I
3647 have made the code more robust (and fixed the comments that describe what
3650 53. Now that there can be additional text after "Completed" in log lines (if
3651 the queue_time_overall log selector is set), a one-byte patch to exigrep
3652 was needed to allow it to recognize "Completed" as not the last thing in
3655 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
3656 patch that reportedly fixes this has been added. I am not expert enough to
3657 create a test for it. This is what the patch creator wrote:
3659 "I found a little strange behaviour of ldap code when working with
3660 Windows 2003 AD Domain, where users was placed in more than one
3661 Organization Units. When I tried to give exim partial DN, the exit code
3662 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
3663 But simultaneously result of request was absolutely normal ldap result,
3664 so I produce this patch..."
3666 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
3667 REFERENCE, so I have modified the code to exclude the patch when that macro
3670 55. Some experimental protocols are using DNS PTR records for new purposes. The
3671 keys for these records are domain names, not reversed IP addresses. The
3672 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3673 leaves it alone. Component reversal etc. now happens only for IP addresses.
3676 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
3678 57. Double the size of the debug message buffer (to 2048) so that more of very
3679 long debug lines gets shown.
3681 58. The exicyclog utility now does better if the number of log files to keep
3682 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
3684 59. Two changes related to the smtp_active_hostname option:
3686 (1) $smtp_active_hostname is now available as a variable.
3687 (2) The default for smtp_banner uses $smtp_active_hostname instead
3688 of $primary_hostname.
3690 60. The host_aton() function is supposed to be passed a string that is known
3691 to be a valid IP address. However, in the case of IPv6 addresses, it was
3692 not checking this. This is a hostage to fortune. Exim now panics and dies
3693 if the condition is not met. A case was found where this could be provoked
3694 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3695 components; fortuitously, this particular loophole had already been fixed
3696 by change 4.50/55 above.
3698 If there are any other similar loopholes, the new check in host_aton()
3699 itself should stop them being exploited. The report I received stated that
3700 data on the command line could provoke the exploit when Exim was running as
3701 exim, but did not say which command line option was involved. All I could
3702 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3703 running as the user.
3706 61. There was a buffer overflow vulnerability in the SPA authentication code
3707 (which came originally from the Samba project). I have added a test to the
3708 spa_base64_to_bits() function which I hope fixes it.
3711 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
3712 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
3714 63. The daemon start-up calls getloadavg() while still root for those OS that
3715 need the first call to be done as root, but it missed one case: when
3716 deliver_queue_load_max is set with deliver_drop_privilege. This is
3717 necessary for the benefit of the queue runner, because there is no re-exec
3718 when deliver_drop_privilege is set.
3720 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
3721 This has been fixed.
3723 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3724 was in use, was not putting the data itself into the right store pool;
3725 consequently, it could be overwritten for a subsequent message in the same
3726 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3729 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
3731 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
3732 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
3733 the function were treating the return as a boolean value, which happened to
3734 work because 0=false and not-0=true, but is not correct code.
3736 68. The host_aton() function was not handling scoped IPv6 addresses (those
3737 with, for example, "%eth0" on the end) correctly.
3739 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
3740 negated items (that is, ~something) in unsigned ints. Some compilers
3741 apparently mutter when there is no cast.
3743 70. If an address verification called from an ACL failed, and did not produce a
3744 user-specific message (i.e. there was only a "system" message), nothing was
3745 put in $acl_verify_message. In this situation, it now puts the system
3748 71. Change 4.23/11 added synchronization checking at the start of an SMTP
3749 session; change 4.31/43 added the unwanted input to the log line - except
3750 that it did not do this in the start of session case. It now does.
3752 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
3753 This is wrong and can cause the other end to generate a synchronization
3754 error if it is another Exim or anything else that does the synchronization
3755 check. A QUIT command is no longer sent after a timeout.
3757 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
3758 during host lookups.
3760 74. The defer_ok option of callout verification was not working if it was used
3761 when verifying addresses in header lines, that is, for this case:
3763 verify = header_sender/callout=defer_ok
3765 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
3766 those file descriptors could be used for SMTP connections. If anything
3767 wrote to stderr (the example that came up was "warn" in embedded Perl), it
3768 could be sent to the SMTP client, causing chaos. The daemon now opens
3769 stdin, stdout, and stderr to /dev/null when it puts itself into the
3772 76. Arrange for output from Perl's "warn" command to be written to Exim's main
3773 log by default. The user can override this with suitable Perl magic.
3775 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
3776 the log message when discard triggers, was not working for the DATA ACL or
3777 for the non-SMTP ACL.
3779 78. Error message wording change in sieve.c.
3781 79. If smtp_accept_max_per_host was set, the number of connections could be
3782 restricted to fewer than expected, because the daemon was trying to set up
3783 a new connection before checking whether the processes handling previous
3784 connections had finished. The check for completed processes is now done
3785 earlier. On busy systems, this bug wouldn't be noticed because something
3786 else would have woken the daemon, and it would have reaped the completed
3789 80. If a message was submitted locally by a user whose login name contained one
3790 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
3791 It caused a spool format error. I have fixed the spool reading code. A
3792 related problem was that the "from" clause in the Received: line became
3793 illegal because of the space(s). It is now covered by ${quote_local_part.
3795 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
3798 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
3800 83. Patch from Sieve maintainer to fix unterminated string problem in
3801 "vacation" handling.
3803 84. Some minor changes to the Linux configuration files to help with other
3804 OS variants using glibc.
3806 85. One more patch for Sieve to update vacation handling to latest spec.
3809 ----------------------------------------------------
3810 See the note above about the 4.44 and 4.50 releases.
3811 ----------------------------------------------------
3817 1. Change 4.43/35 introduced a bug that caused file counts to be
3818 incorrectly computed when quota_filecount was set in an appendfile
3821 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3822 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3825 3. Give more explanation in the error message when the command for a transport
3826 filter fails to execute.
3828 4. There are several places where Exim runs a non-Exim command in a
3829 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3830 was being done only for the command run by the queryprogram router. It is
3831 now done for all such subprocesses. The other cases are: ${run, transport
3832 filters, and the commands run by the lmtp and pipe transports.
3834 5. Some older OS have a limit of 256 on the maximum number of file
3835 descriptors. Exim was using setrlimit() to set 1000 as a large value
3836 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3837 systems. I've change it so that if it can't get 1000, it tries for 256.
3839 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3840 was an oversight, and furthermore, ever since the addition of extra
3841 controls (e.g. 4.43/32), the checks on when to allow different forms of
3842 "control" were broken. There should now be diagnostics for all cases when a
3843 control that does not make sense is encountered.
3845 7. $recipients is now available in the predata ACL (oversight).
3847 8. Tidy the search cache before the fork to do a delivery from a message
3848 received from the command line. Otherwise the child will trigger a lookup
3849 failure and thereby defer the delivery if it tries to use (for example) a
3850 cached ldap connection that the parent has called unbind on.
3852 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3853 of $address_data from the recipient verification was clobbered by the
3854 sender verification.
3856 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3857 was its contents. (It was OK if the option was not defined at all.)
3859 11. A "Completed" log line is now written for messages that are removed from
3860 the spool by the -Mrm option.
3862 12. $host_address is now set to the target address during the checking of
3863 ignore_target_hosts.
3865 13. When checking ignore_target_hosts for an ipliteral router, no host name was
3866 being passed; this would have caused $sender_host_name to have been used if
3867 matching the list had actually called for a host name (not very likely,
3868 since this list is usually IP addresses). A host name is now passed as
3871 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3872 code that specifies a non-restarting handler (typically sigaction() in
3873 modern systems) in an attempt to fix a rare and obscure crash bug.
3875 15. Narrowed the window for a race in the daemon that could cause it to ignore
3876 SIGCHLD signals. This is not a major problem, because they are used only to
3877 wake it up if nothing else does.
3879 16. A malformed maildirsize file could cause Exim to calculate negative values
3880 for the mailbox size or file count. Odd effects could occur as a result.
3881 The maildirsize information is now recalculated if the size or filecount
3884 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3885 support for a long time. Removed HAVE_SYS_VFS_H.
3887 18. Updated exipick to current release from John Jetmore.
3889 19. Allow an empty sender to be matched against a lookup in an address list.
3890 Previously the only cases considered were a regular expression, or an
3893 20. Exim went into a mad DNS lookup loop when doing a callout where the
3894 host was specified on the transport, if the DNS lookup yielded more than
3897 21. The RFC2047 encoding function was originally intended for short strings
3898 such as real names; it was not keeping to the 75-character limit for
3899 encoded words that the RFC imposes. It now respects the limit, and
3900 generates multiple encoded words if necessary. To be on the safe side, I
3901 have increased the buffer size for the ${rfc2047: expansion operator from
3904 22. Failure to deliver a bounce message always caused it to be frozen, even if
3905 there was an errors_to setting on the router. The errors_to setting is now
3908 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3909 canonical form (fully expanded) before being placed in
3910 $sender_host_address.
3912 24. Updated eximstats to version 1.33
3914 25. Include certificate and key file names in error message when GnuTLS fails
3915 to set them up, because the GnuTLS error message doesn't include the name
3916 of the failing file when there is a problem reading it.
3918 26. Expand error message when OpenSSL has problems setting up cert/key files.
3921 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
3922 (this can affect the format of dates).
3924 28. exim_tidydb, when checking for the continued existence of a message for
3925 which it has found a message-specific retry record, was not finding
3926 messages that were in split spool directories. Consequently, it was
3927 deleting retry records that should have stayed in existence.
3929 29. eximstats updated to version 1.35
3930 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
3931 1.35 - bugfix such that pie charts by volume are generated correctly
3933 30. The SPA authentication driver was not abandoning authentication and moving
3934 on to the next authenticator when an expansion was forced to fail,
3935 contradicting the general specification for all authenticators. Instead it
3936 was generating a temporary error. It now behaves as specified.
3938 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
3939 (the order specifies the preference for clients). The order is now AES256,
3940 AES128, 3DES, ARCFOUR128.
3942 31. Small patch to Sieve code - explicitly set From: when generating an
3945 32. Exim crashed if a remote delivery caused a very long error message to be
3946 recorded - for instance if somebody sent an entire SpamAssassin report back
3947 as a large number of 550 error lines. This bug was coincidentally fixed by
3948 increasing the size of one of Exim's internal buffers (big_buffer) that
3949 happened as part of the Exiscan merge. However, to be on the safe side, I
3950 have made the code more robust (and fixed the comments that describe what
3953 33. Some experimental protocols are using DNS PTR records for new purposes. The
3954 keys for these records are domain names, not reversed IP addresses. The
3955 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3956 leaves it alone. Component reversal etc. now happens only for IP addresses.
3959 34. The host_aton() function is supposed to be passed a string that is known
3960 to be a valid IP address. However, in the case of IPv6 addresses, it was
3961 not checking this. This is a hostage to fortune. Exim now panics and dies
3962 if the condition is not met. A case was found where this could be provoked
3963 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3964 components; fortuitously, this particular loophole had already been fixed
3965 by change 4.50/55 or 4.44/33 above.
3967 If there are any other similar loopholes, the new check in host_aton()
3968 itself should stop them being exploited. The report I received stated that
3969 data on the command line could provoke the exploit when Exim was running as
3970 exim, but did not say which command line option was involved. All I could
3971 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3972 running as the user.
3975 35. There was a buffer overflow vulnerability in the SPA authentication code
3976 (which came originally from the Samba project). I have added a test to the
3977 spa_base64_to_bits() function which I hope fixes it.
3980 36. The daemon start-up calls getloadavg() while still root for those OS that
3981 need the first call to be done as root, but it missed one case: when
3982 deliver_queue_load_max is set with deliver_drop_privilege. This is
3983 necessary for the benefit of the queue runner, because there is no re-exec
3984 when deliver_drop_privilege is set.
3986 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3987 was in use, was not putting the data itself into the right store pool;
3988 consequently, it could be overwritten for a subsequent message in the same
3989 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3992 38. Sometimes the final signoff response after QUIT could fail to get
3993 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3994 < 0 before doing a fflush(). This bug looks as though it goes back to the
3995 introduction of TLS in release 3.20, but "sometimes" must have been rare
3996 because the tests only now provoked it.
4002 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4003 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4004 Now it returns an integer. A number of calls were still expecting a T/F
4005 return. Fortuitously, in all cases, the tests worked in OK situations,
4006 which is the norm. However, things would have gone wrong on any write
4007 failures on the smtp file descriptor. This function is used when sending
4008 messages over SMTP and also when doing verify callouts.
4010 2. When Exim is called to do synchronous delivery of a locally submitted
4011 message (the -odf or -odi options), it no longer closes stderr before doing
4014 3. Implemented the mua_wrapper option.
4016 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4018 5. Implemented the functions header_remove(), header_testname(),
4019 header_add_at_position(), and receive_remove_recipient(), and exported them
4022 6. If an ACL "warn" statement specified the addition of headers, Exim already
4023 inserted X-ACL-Warn: at the start if there was no header name. However, it
4024 was not making this test for the second and subsequent header lines if
4025 there were newlines in the string. This meant that an invalid header could
4026 be inserted if Exim was badly configured.
4028 7. Allow an ACL "warn" statement to add header lines at the start or after all
4029 the Received: headers, as well as at the end.
4031 8. Added the rcpt_4xx retry error code.
4033 9. Added postmaster_mailfrom=xxx to callout verification option.
4035 10. Added mailfrom=xxxx to the callout verification option, for verify=
4038 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4039 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4041 12. Inserted some casts to stop certain compilers warning when using pointer
4042 differences as field lengths or precisions in printf-type calls (mostly
4043 affecting debugging statements).
4045 13. Added optional readline() support for -be (dynamically loaded).
4047 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4048 same clock tick as a message's arrival, so that its received time was the
4049 same as the "first fail" time on the retry record, and that message
4050 remained on the queue past the ultimate address timeout, every queue runner
4051 would try a delivery (because it was past the ultimate address timeout) but
4052 after another failure, the ultimate address timeout, which should have then
4053 bounced the address, did not kick in. This was a "< instead of <=" error;
4054 in most cases the first failure would have been in the next clock tick
4055 after the received time, and all would be well.
4057 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4058 being recognized when the domain list was tested by the match_domain
4059 condition in an expansion string.
4061 16. Added the ${str2b64: operator.
4063 17. Exim was always calling setrlimit() to set a large limit for the number of
4064 processes, without checking whether the existing limit was already
4065 adequate. (It did check for the limit on file descriptors.) Furthermore,
4066 errors from getrlimit() and setrlimit() were being ignored. Now they are
4067 logged to the main and panic logs, but Exim does carry on, to try to do its
4068 job under whatever limits there are.
4070 18. Imported PCRE 5.0.
4072 19. Trivial typo in log message " temporarily refused connection" (the leading
4075 20. If the log selector return_path_on_delivery was set and an address was
4076 redirected to /dev/null, the delivery process crashed because it assumed
4077 that a return path would always be set for a "successful" delivery. In this
4078 case, the whole delivery is bypassed as an optimization, and therefore no
4081 21. Internal re-arrangement: the function for sending a challenge and reading
4082 a response while authentication was assuming a zero-terminated challenge
4083 string. It's now changed to take a pointer and a length, to allow for
4084 binary data in such strings.
4086 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4088 23. Exim was not respecting finduser_retries when seeking the login of the
4089 uid under which it was called; it was always trying 10 times. (The default
4090 setting of finduser_retries is zero.) Also, it was sleeping after the final
4091 failure, which is pointless.
4093 24. Implemented tls_on_connect_ports.
4095 25. Implemented acl_smtp_predata.
4097 26. If the domain in control=submission is set empty, Exim assumes that the
4098 authenticated id is a complete email address when it generates From: or
4099 Sender: header lines.
4101 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4102 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4103 chown and chgrp in /bin and hostname in /usr/bin.
4105 28. Exim was keeping the "process log" file open after each use, just as it
4106 does for the main log. This opens the possibility of it remaining open for
4107 long periods when the USR1 signal hits a daemon. Occasional processlog
4108 errors were reported, that could have been caused by this. Anyway, it seems
4109 much more sensible not to leave this file open at all, so that is what now
4112 29. The long-running daemon process does not normally write to the log once it
4113 has entered its main loop, and it closes the log before doing so. This is
4114 so that log files can straightforwardly be renamed and moved. However,
4115 there are a couple of unusual error situations where the daemon does write
4116 log entries, and I had neglected to close the log afterwards.
4118 30. The text of an SMTP error response that was received during a remote
4119 delivery was being truncated at 512 bytes. This is too short for some of
4120 the long messages that one sometimes sees. I've increased the limit to
4123 31. It is now possible to make retry rules that apply only when a message has a
4124 specific sender, in particular, an empty sender.
4126 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4127 it possible to be selective about when SMTP synchronization is enforced.
4129 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4131 32. Implemented hosts_connection_nolog.
4133 33. Added an ACL for QUIT.
4135 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4138 35. Added mailbox_size and mailbox_filecount to appendfile.
4140 36. Added control = no_multiline_responses to ACLs.
4142 37. There was a bug in the logic of the code that waits for the clock to tick
4143 in the case where the clock went backwards by a substantial amount such
4144 that the microsecond fraction of "now" was more than the microsecond
4145 fraction of "then" (but the whole seconds number was less).
4147 38. Added support for the libradius Radius client library this is found on
4148 FreeBSD (previously only the radiusclient library was supported).
4154 1. When certain lookups returned multiple values in the form name=value, the
4155 quoting of the values was not always being done properly. Specifically:
4156 (a) If the value started with a double quote, but contained no whitespace,
4158 (b) If the value contained whitespace other than a space character (i.e.
4159 tabs or newlines or carriage returns) it was not quoted.
4160 This fix has been applied to the mysql and pgsql lookups by writing a
4161 separate quoting function and calling it from the lookup code. The fix
4162 should probably also be applied to nisplus, ibase and oracle lookups, but
4163 since I cannot test any of those, I have not disturbed their existing code.
4165 2. A hit in the callout cache for a specific address caused a log line with no
4166 reason for rejecting RCPT. Now it says "Previous (cached) callout
4167 verification failure".
4169 3. There was an off-by-one bug in the queryprogram router. An over-long
4170 return line was truncated at 256 instead of 255 characters, thereby
4171 overflowing its buffer with the terminating zero. As well as fixing this, I
4172 have increased the buffer size to 1024 (and made a note to document this).
4174 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4175 when Exim is waiting for an SMTP response from a remote server, Exim
4176 restarts its select() call on the socket, thereby resetting its timeout.
4177 This is not a problem when such interrupts are rare. Somebody set up a cron
4178 job to run exiwhat every 2 minutes, which is less than the normal select()
4179 timeout (5 or 10 minutes). This meant that the select() timeout never
4180 kicked in because it was always reset. I have fixed this by comparing the
4181 time when an interrupt arrives with the time at the start of the first call
4182 to select(). If more time than the timeout has elapsed, the interrupt is
4183 treated as a timeout.
4185 5. Some internal re-factoring in preparation for the addition of Sieve
4186 extensions (by MH). In particular, the "personal" test is moved to a
4187 separate function, and given an option for scanning Cc: and Bcc: (which is
4188 not set for Exim filters).
4190 6. When Exim created an email address using the login of the caller as the
4191 local part (e.g. when creating a From: or Sender: header line), it was not
4192 quoting the local part when it contained special characters such as @.
4194 7. Installed new OpenBSD configuration files.
4196 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4197 try to make them clearer.
4199 9. Callout options, other than the timeout value, were being ignored when
4200 verifying sender addresses in header lines. For example, when using
4202 verify = header_sender/callout=no_cache
4204 the cache was (incorrectly) being used.
4206 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4207 only the Cygwin environment.
4209 11. When return_path_on_delivery was set as a log selector, if different remote
4210 addresses in the same message used different return paths and parallel
4211 remote delivery occurred, the wrong values would sometimes be logged.
4212 (Whenever a remote delivery process finished, the return path value from
4213 the most recently started remote delivery process was logged.)
4215 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4216 lines when AUTH and/or TLS are in use. This is the "received protocol"
4217 field. Exim used to use "asmtp" for authenticated SMTP, without any
4218 indication (in the protocol name) for TLS use. Now it follows the RFC and
4219 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4220 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4221 names appear in log lines as well as in Received: header lines.
4223 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4224 extensions, and comparison tests, and to fix some bugs.
4226 14. Changes to the "personal" filter test:
4228 (1) The test was buggy in that it was just doing the equivalent of
4229 "contains" tests on header lines. For example, if a user's address was
4230 anne@some.where, the "personal" test would incorrectly be true for
4232 To: susanne@some.where
4234 This test is now done by extracting each address from the header in turn,
4235 and checking the entire address. Other tests that are part of "personal"
4236 are now done using regular expressions (for example, to check local parts
4237 of addresses in From: header lines).
4239 (2) The list of non-personal local parts in From: addresses has been
4240 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4241 taken from the Sieve specification recommendations.
4243 (3) If the message contains any header line starting with "List-" it is
4244 treated as non-personal.
4246 (4) The test for "circular" in the Subject: header line has been removed
4247 because it now seems ill-conceived.
4249 15. Minor typos in src/EDITME comments corrected.
4251 16. Installed latest exipick from John Jetmore.
4253 17. If headers_add on a router specified a text string that was too long for
4254 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4255 of string_sprintf() is now avoided.
4257 18. $message_body_size was not set (it was always zero) when running the DATA
4258 ACL and the local_scan() function.
4260 19. For the "mail" command in an Exim filter, no default was being set for
4261 the once_repeat time, causing a random time value to be used if "once" was
4262 specified. (If the value happened to be <= 0, no repeat happened.) The
4263 default is now 0s, meaning "never repeat". The "vacation" command was OK
4264 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4265 (I found it when inspecting the code).
4267 20. There is now an overall timeout for performing a callout verification. It
4268 defaults to 4 times the callout timeout, which applies to individual SMTP
4269 commands during the callout. The overall timeout applies when there is more
4270 than one host that can be tried. The timeout is checked before trying the
4271 next host. This prevents very long delays if there are a large number of
4272 hosts and all are timing out (e.g. when the network connections are timing
4273 out). The value of the overall timeout can be changed by specifying an
4274 additional sub-option for "callout", called "maxwait". For example:
4276 verify = sender/callout=5s,maxwait=20s
4278 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4279 to the end before writing, but this should make it even safer).
4281 22. Exim was forgetting that it had advertised PIPELINING for the second and
4282 subsequent messages on an SMTP connection. It was also not resetting its
4283 memory on STARTTLS and an internal HELO.
4285 23. When Exim logs an SMTP synchronization error within a session, it now
4286 records whether PIPELINING has been advertised or not.
4288 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4289 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4290 rather than long int.
4292 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4294 26. Added the never_mail option to autoreply.
4300 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4301 crash if the getsockname() call failed; this can happen if a connection is
4302 closed very soon after it is established. The problem was simply in the
4303 order in which certain operations were done, causing Exim to try to write
4304 to the SMTP stream before it had set up the file descriptor. The bug has
4305 been fixed by making things happen in the correct order.
4311 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4312 before the connection was closed, thus losing the rejection response.
4314 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4315 some early Solaris releases, but causes trouble in current releases where
4316 socklen_t is defined.
4318 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4321 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4324 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4325 that releasing the top of it at the end releases what was used for sub-
4326 expansions (unless the block got too big). However, discard this block if
4327 the first thing is a variable or header, so that we can use its block when
4328 it is dynamic (useful for very large $message_headers, for example).
4330 6. Lookups now cache *every* query, not just the most recent. A new, separate
4331 store pool is used for this. It can be recovered when all lookup caches are
4332 flushed. Lookups now release memory at the end of their result strings.
4333 This has involved some general refactoring of the lookup sources.
4335 7. Some code has been added to the store_xxx() functions to reduce the amount
4336 of flapping under certain conditions.
4338 8. log_incoming_interface used to affect only the <= reception log lines. Now
4339 it causes the local interface and port to be added to several more SMTP log
4340 lines, for example "SMTP connection from", and rejection lines.
4342 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4344 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4346 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4347 could be overwritten at the end of the current message (or the start of a
4348 new message if it was set in a HELO ACL). The value is now preserved for
4349 the duration of the SMTP connection.
4351 12. If a transport had a headers_rewrite setting, and a matching header line
4352 contained an unqualified address, that address was qualified, even if it
4353 did not match any rewriting rules. The underlying bug was that the values
4354 of the flags that permit the existence of unqualified sender and recipient
4355 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4356 non-local messages, and by -bnq for local messages) were not being
4357 preserved with the message after it was received.
4359 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4360 "next input=" as part of the text comprising the host identity instead of
4361 the correct text. The code was using the same buffer for two different
4362 strings. However, depending on which order the printing function evaluated
4363 its arguments, the bug did not always show up. Under Linux, for example, my
4364 test suite worked just fine.
4366 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4367 doesn't work with some older versions of Perl. It has been changed to "my",
4368 which in any case is probably the better facility to use.
4370 15. A really picky compiler found some instances of statements for creating
4371 error messages that either had too many or two few arguments for the format
4374 16. The size of the buffer for calls to the DNS resolver has been increased
4375 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4376 for addresses that have a lot of PTR records. This alleviates a problem; it
4377 does not fully solve it.
4379 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4380 buffer now truncates the list and logs the incident, which is the same
4381 action as happens when Exim is looking up a host name and its aliases.
4382 Previously in this situation something unpredictable would happen;
4383 sometimes it was "internal error: store_reset failed".
4385 18. If a server dropped the connection unexpectedly when an Exim client was
4386 using GnuTLS and trying to read a response, the client delivery process
4387 crashed while trying to generate an error log message.
4389 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4390 string, for example:
4392 warn message = H1: something\nH2: something
4394 the text was added as a single header line from Exim's point of view
4395 though it ended up OK in the delivered message. However, searching for the
4396 second and subsequent header lines using $h_h2: did not work. This has been
4397 fixed. Similarly, if a system filter added multiple headers in this way,
4398 the routers could not see them.
4400 20. Expanded the error message when iplsearch is called with an invalid key to
4401 suggest using net-iplsearch in a host list.
4403 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4404 ACLs are no longer actually imposed (and a message to that effect is
4407 22. If a "gecos" field in a passwd entry contained escaped characters, in
4408 particular, if it contained a \" sequence, Exim got it wrong when building
4409 a From: or a Sender: header from that name. A second bug also caused
4410 incorrect handling when an unquoted " was present following a character
4411 that needed quoting.
4413 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4414 was not being matched caselessly.
4416 24. Arranged for all hyphens in the exim.8 source to be escaped with
4419 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4420 properly. Recipient callout cache records were still being keyed to include
4421 the sender, even when use_sender was set false. This led to far more
4422 callouts that were necessary. The sender is no longer included in the key
4423 when use_sender is false.
4425 26. Added "control = submission" modifier to ACLs.
4427 27. Added the ${base62d: operator to decode base 62 numbers.
4429 28. dnsdb lookups can now access SRV records.
4431 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4432 the configuration file.
4434 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4435 (-v) mode. This makes the output for a verbose queue run more intelligible.
4437 31. Added a use_postmaster feature to recipient callouts.
4439 32. Added the $body_zerocount variable, containing the number of binary zero
4440 bytes in the message body.
4442 33. The time of last modification of the "new" subdirectory is now used as the
4443 "mailbox time last read" when there is a quota error for a maildir
4446 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4448 35. Added +ignore_unknown as a special item in host lists.
4450 36. Code for decoding IPv6 addresses in host lists is now included, even if
4451 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4452 address was recognized as an IP address, but was then not correctly decoded
4453 into binary, causing unexpected and incorrect effects when compared with
4460 1. Very minor rewording of debugging text in manualroute to say "list of
4461 hosts" instead of "hostlist".
4463 2. If verify=header_syntax was set, and a header line with an unqualified
4464 address (no domain) and a large number of spaces between the end of the
4465 name and the colon was received, the reception process suffered a buffer
4466 overflow, and (when I tested it) crashed. This was caused by some obsolete
4467 code that should have been removed. The fix is to remove it!
4469 3. When running in the test harness, delay a bit after writing a bounce
4470 message to get a bit more predictability in the log output.
4472 4. Added a call to search_tidyup() just before forking a reception process. In
4473 theory, someone could use a lookup in the expansion of smtp_accept_max_
4474 per_host which, without the tidyup, could leave open a database connection.
4476 5. Added the variables $recipient_data and $sender_data which get set from a
4477 lookup success in an ACL "recipients" or "senders" condition, or a router
4478 "senders" option, similar to $domain_data and $local_part_data.
4480 6. Moved the writing of debug_print from before to after the "senders" test
4483 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4484 problems for message scanning, either using a data ACL, or using
4485 local_scan() because the Received: header was not generated till after they
4486 were called (in order to set the time as the time of reception completion).
4487 I have revised the way this works. The header is now generated after the
4488 body is received, but before the ACL or local_scan() are called. After they
4489 are run, the timestamp in the header is updated.
4495 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4496 before starting a queue runner without re-exec. This happened only when
4497 deliver_drop_privilege was set or when the Exim user was set to root. The
4498 effect of the bug was that timeouts during subsequent deliveries caused
4499 crashes instead of being properly handled. The handler is now left at its
4500 default (and expected) setting.
4502 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4503 message, again when deliver_drop_privilege is set or Exim is run as root.
4504 The bug described in (1) was not present in this case, but the tidying up
4505 of the other signals was missing. I have made the two cases consistent.
4507 3. The ignore_target_hosts setting on a manualroute router was being ignored
4508 for hosts that were looked up using the /MX notation.
4510 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4513 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4514 operated on the sender address. After changing the $sender_address to <>
4515 for the sender address verify, Exim was re-instated it as the original
4516 (before rewriting) address, but remembering that it had rewritten it, so it
4517 wasn't rewriting it again. This bug also had the effect of breaking the
4518 sender address verification caching when the sender address was rewritten.
4520 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4521 This has been changed so that if the ip literal address matches
4522 ignore_target_hosts, the router declines.
4524 7. Added expansion conditions match_domain, match_address, and match_local_
4525 part (NOT match_host).
4527 8. The placeholder for the Received: header didn't have a length field set.
4529 9. Added code to Exim itself and to exim_lock to test for a specific race
4530 condition that could lead to file corruption when using MBX delivery. The
4531 issue is with the lockfile that is created in /tmp. If this file is removed
4532 after a process has opened it but before that process has acquired a lock,
4533 there is the potential for a second process to recreate the file and also
4534 acquire a lock. This could lead to two Exim processes writing to the file
4535 at the same time. The added code performs the same test as UW imapd; it
4536 checks after acquiring the lock that its file descriptor still refers to
4537 the same named file.
4539 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4540 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4543 11. Added the smtp_active_hostname option. If used, this will typically be made
4544 to depend on the incoming interface address. Because $interface_address is
4545 not set up until the daemon has forked a reception process, error responses
4546 that can happen earlier (such as "too many connections") no longer contain
4549 12. If an expansion in a condition on a "warn" statement fails because a lookup
4550 defers, the "warn" statement is abandoned, and the next ACL statement is
4551 processed. Previously this caused the whole ACL to be aborted.
4553 13. Added the iplsearch lookup type.
4555 14. Added ident_timeout as a log selector.
4557 15. Added tls_certificate_verified as a log selector.
4559 16. Added a global option tls_require_ciphers (compare the smtp transport
4560 option of the same name). This controls incoming TLS connections.
4562 17. I finally figured out how to make tls_require_ciphers do a similar thing
4563 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
4564 before starting the TLS session.
4566 18. Tabs are now shown as \t in -bP output.
4568 19. If the log selector return_path_on_delivery was set, Exim crashed when
4569 bouncing a message because it had too many Received: header lines.
4571 20. If two routers both had headers_remove settings, and the first one included
4572 a superfluous trailing colon, the final name in the first list and the
4573 first name in the second list were incorrectly joined into one item (with a
4574 colon in the middle).
4580 1. Added -C and -D options to the exinext utility, mainly to make it easier
4581 to include in the automated testing, but these could be helpful when
4582 multiple configurations are in use.
4584 2. The exinext utility was not formatting the output nicely when there was
4585 an alternate port involved in the retry record key, nor when there was a
4586 message id as well (for retries that were specific to a specific message
4587 and a specific host). It was also confused by IPv6 addresses, because of
4588 the additional colons they contain. I have fixed the IPv4 problem, and
4589 patched it up to do a reasonable job for IPv6.
4591 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
4592 delivery, the log line now contains "pipelined" if PIPELINING was used.
4594 4. An SMTP transport process used to panic and die if the bind() call to set
4595 an explicit outgoing interface failed. This has been changed; it is now
4596 treated in the same way as a connect() failure.
4598 5. A reference to $sender_host_name in the part of a conditional expansion
4599 that was being skipped was still causing a DNS lookup. This no longer
4602 6. The def: expansion condition was not recognizing references to header lines
4603 that used bh_ and bheader_.
4605 7. Added the _cache feature to named lists.
4607 8. The code for checking quota_filecount in the appendfile transport was
4608 allowing one more file than it should have been.
4610 9. For compatibility with Sendmail, the command line option
4618 and sets the incoming protocol and host name (for trusted callers). The
4619 host name and its colon can be omitted when only the protocol is to be set.
4620 Note the Exim already has two private options, -pd and -ps, that refer to
4621 embedded Perl. It is therefore impossible to set a protocol value of "d" or
4622 "s", but I don't think that's a major issue.
4624 10. A number of refactoring changes to the code, none of which should affect
4627 (a) The number of logging options was getting close to filling up the
4628 32-bit word that was used as a bit map. I have split them into two classes:
4629 those that are passed in the argument to log_write(), and those that are
4630 only ever tested independently outside of that function. These are now in
4631 separate 32-bit words, so there is plenty of room for expansion again.
4632 There is no change in the user interface or the logging behaviour.
4634 (b) When building, for example, log lines, the code previously used a
4635 macro that called string_cat() twice, in order to add two strings. This is
4636 not really sufficiently general. Furthermore, there was one instance where
4637 it was actually wrong because one of the argument was used twice, and in
4638 one call a function was used. (As it happened, calling the function twice
4639 did not affect the overall behaviour.) The macro has been replaced by a
4640 function that can join an arbitrary number of extra strings onto a growing
4643 (c) The code for expansion conditions now uses a table and a binary chop
4644 instead of a serial search (which was left over from when there were very
4645 few conditions). Also, it now recognizes conditions like "pam" even when
4646 the relevant support is not compiled in: a suitably worded error message is
4647 given if an attempt is made to use such a condition.
4649 11. Added ${time_interval:xxxxx}.
4651 12. A bug was causing one of the ddress fields not to be passed back correctly
4652 from remote delivery subprocesses. The field in question was not being
4653 subsequently used, so this caused to problems in practice.
4655 13. Added new log selectors queue_time and deliver_time.
4657 14. Might have fixed a bug in maildirsizefile handling that threw up
4658 "unexpected character" debug warnings, and recalculated the data
4659 unnecessarily. In any case, I expanded the warning message to give more
4662 15. Added the message "Restricted characters in address" to the statements in
4663 the default ACL that block characters like @ and % in local parts.
4665 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
4666 Three changes have been made:
4668 (a) There was a serious bug; a negative response to MAIL caused the whole
4669 recipient domain to be cached as invalid, thereby blocking all messages
4670 to all local parts at the same domain, from all senders. This bug has
4671 been fixed. The domain is no longer cached after a negative response to
4672 MAIL if the sender used is not empty.
4674 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
4677 (c) A new callout option, "use_sender" has been added for people who want
4678 the modified behaviour.
4684 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
4687 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
4688 indeed breaks things for older releases.
4690 3. Added additional logging to the case where there is a problem reading data
4691 from a filter that is running in a subprocess using a pipe, in order to
4692 try to track down a specific problem.
4694 4. Testing facility fudge: when running in the test harness and attempting
4695 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
4696 getting "No route to host". Convert this to a timeout.
4698 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
4701 6. Some OS don't have socklen_t but use size_t instead. This affects the
4702 fifth argument of getsockopt() amongst other things. This is now
4703 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
4704 can be set for individual OS. I have set it for SunOS5, OSF1, and
4705 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
4706 some earlier ones do not.
4708 7. Change 4.30/15 was not doing the test caselessly.
4710 8. The standard form for an IPv6 address literal was being rejected by address
4711 parsing in, for example, MAIL and RCPT commands. An example of this kind of
4712 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4713 this, as well as the form without the "IPv6" on the front (but only when
4714 address literals are enabled, of course).
4716 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
4718 10. Exim crashed if a message with an empty sender address specified by -f
4719 encountered a router with an errors_to setting. This could be provoked only
4720 by a command such as
4724 where an empty string was supplied; "<>" did not hit this bug.
4726 11. Installed PCRE release 4.5.
4728 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
4729 remained set. It is now erased.
4731 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
4732 times from message ids (which are base 36 rather than the normal 62).
4734 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
4735 were being counted as actual protocol errors, and logged if the log
4736 selector +smtp_protocol_error was set. One cannot be perfect in this test,
4737 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
4738 and DATA following a set of rejected RCPTs do not count as protocol errors.
4739 In other words, Exim assumes they were pipelined, though this may not
4740 actually be the case. Of course, in all cases the client gets an
4741 appropriate error code.
4743 15. If a lookup fails in an ACL condition, a message about the failure may
4744 be available; it is used if testing the ACL cannot continue, because most
4745 such messages specify what the cause of the deferral is. However, some
4746 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
4747 that caused an old message to be retained and used if a later statement
4748 caused a defer, replacing the real cause of the deferral.
4750 16. If an IP address had so many PTR records that the DNS lookup buffer
4751 was not large enough to hold them, Exim could crash while trying to process
4752 the truncated data. It now detects and logs this case.
4754 17. Further to 4.21/58, another change has been made: if (and only if) the
4755 first line of a message (the first header line) ends with CRLF, a bare LF
4756 in a subsequent header line has a space inserted after it, so as not to
4757 terminate the header.
4759 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
4760 unnecessarily, used atoi() instead of strtol(), and didn't check the
4761 termination when getting file sizes from file names by regex.
4763 19. Completely re-implemented the support for maildirsize files, in the light
4764 of a number of problems with the previous contributed implementation
4765 (4.30/29). In particular:
4767 . If the quota is zero, the maildirsize file is maintained, but no quota is
4770 . If the maildir directory does not exist, it is created before any attempt
4771 to write a maildirsize file.
4773 . The quota value in the file is just a cache; if the quota is changed in
4774 the transport, the new value overrides.
4776 . A regular expression is available for excluding directories from the
4779 20. The autoreply transport checks the characters in options that define the
4780 message's headers; it allows continued headers, but it was checking with
4781 isspace() after an embedded newline instead of explicitly looking for a
4784 21. If all the "regular" hosts to which an address was routed had passed their
4785 expiry times, and had not reached their retry times, the address was
4786 bounced, even if fallback hosts were defined. Now Exim should go on to try
4789 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
4790 equivalent code in the SMTP transport. Some hosts send humungous responses
4791 to HELO/EHLO, more than 1024 it seems.
4793 23. Refactoring: code in filter.c used (void *) for "any old type" but this
4794 gives compiler warnings in some environments. I've now done it "properly",
4797 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
4798 (because of problems with the built-in one) was declared to return uschar *
4799 instead of char *, causing compiler failure.
4801 25. Fixed a file descriptor leak when processing alias/forward files.
4803 26. Fixed a minor format string issue in dbfn.c.
4805 27. Typo in exim.c: ("dmbnz" for "dbmnz").
4807 28. If a filter file refered to $h_xxx or $message_headers, and the headers
4808 contained RFC 2047 "words", Exim's memory could, under certain conditions,
4811 29. When a sender address is verified, it is cached, to save repeating the test
4812 when there is more than one recipient in a message. However, when the
4813 verification involves a callout, it is possible for different callout
4814 options to be set for different recipients. It is too complicated to keep
4815 track of this in the cache, so now Exim always runs a verification when a
4816 callout is required, relying on the callout cache for the optimization.
4817 The overhead is duplication of the address routing, but this should not be
4820 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
4821 to be verified with callout=postmaster, and the main callout worked but the
4822 postmaster check failed, the verification correctly failed. However, if a
4823 subsequent RCPT command asked for sender verification *without* the
4824 postmaster check, incorrect caching caused this verification also to fail,
4827 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
4828 it was not caching the DNS options (qualify_single, search_parents) that
4829 were used when the lookup failed. A subsequent lookup with different
4830 options therefore always gave the same answer, though there were cases
4831 where it should not have. (Example: a "domains = !$mx_any" option on a
4832 dnslookup router: the "domains" option is always processed without any
4833 widening, but the router might have qualify_single set.) Now Exim uses the
4834 cached value only when the same options are set.
4836 32. Added John Jetmore's "exipick" utility to the distribution.
4838 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
4839 than a timeout (e.g. a certificate is required, and is not provided), an
4840 Exim server now closes the connection immediately. Previously it waited for
4841 the client to close - but if the client is SSL, it seems that they each
4842 wait for each other, leading to a delay before one of them times out.
4844 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
4845 maintained 0.8.x compatibility because I don't think many are using it, and
4846 it is clearly obsolete.
4848 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
4851 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
4852 client certificate was expired. A simple patch fixes this, though I don't
4853 understand the full logic of why the verify callback is called multiple
4856 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
4857 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
4858 which causes problems with some clients (such as the Certicom SSL Plus
4859 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
4860 disables the coutermeasure allowing Eudora to connect."
4862 38. Exim was not checking that a write() to a log file succeeded. This could
4863 lead to Bad Things if a log got too big, in particular if it hit a file
4864 size limit. Exim now panics and dies if it cannot write to a log file, just
4865 as it does if it cannot open a log file.
4867 39. Modified OS/Makefile-Linux so that it now contains
4869 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
4871 The two -D definitions ensure that Exim is compiled with large file
4872 support, which makes it possible to handle log files that are bigger than
4875 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
4876 instance) a domain was checked against a named list that involved a lookup,
4877 causing $domain_data to be set, then another domain was checked against the
4878 same list, then the first domain was re-checked, the value of $domain_data
4879 after the final check could be wrong. In particular, if the second check
4880 failed, it could be set empty. This bug probably also applied to
4883 41. The strip_trailing_dot option was not being applied to the address given
4884 with the -f command-line option.
4886 42. The code for reading a message's header from the spool was incrementing
4887 $received_count, but never initializing it. This meant that the value was
4888 incorrect (doubled) while delivering a message in the same process in which
4889 it was received. In the most common configuration of Exim, this never
4890 happens - a fresh exec is done - but it can happen when
4891 deliver_drop_privilege is set.
4893 43. When Exim logs an SMTP synchronization error - client data sent too soon -
4894 it now includes up to 150 characters of the unexpected data in the log
4897 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
4898 and building data strings. The size of both of these buffers was 10 000
4899 bytes - far larger than anybody would *ever* want, thought I. Needless to
4900 say, somebody hit the limit. I have increased the maximum line length to
4901 20 000 and the maximum data length of concatenated lines to 100 000. I have
4902 also fixed two bugs, because there was no checking on these buffers. Tsk,
4903 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
4904 buffer is too small.
4906 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
4907 lsearch lookups. Now it does.
4909 46. When parsing a route_list item in a manualroute router, a fixed-length
4910 buffer was used for the list of hosts. I made this 1024 bytes long,
4911 thinking that nobody would ever have a list of hosts that long. Wrong.
4912 Somebody had a whole pile of complicated expansion conditions, and the
4913 string was silently truncated, leading to an expansion error. It turns out
4914 that it is easier to change to an unlimited length (owing to other changes
4915 that have happened since this code was originally written) than to build
4916 structure for giving a limitation error. The length of the item that
4917 expands into the list of hosts is now unlimited.
4919 47. The lsearch lookup could not handle data where the length of text line was
4920 more than 4095 characters. Such lines were truncated, leading to shortened
4921 data being returned. It should now handle lines of any length.
4923 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
4924 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
4927 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
4928 replacing strings such as BIN_DIRECTORY in a source file: the replacement
4929 no longer happens in comment lines. A list of replacements is now placed
4930 at the head of all of the source files, except those whose only change is
4931 to replace PERL_COMMAND in the very first #! line.
4933 50. Replaced the slow insertion sort in queue.c, for sorting the list of
4934 messages on the queue, with a bottom-up merge sort, using code contributed
4935 by Michael Haardt. This should make operations like -bp somewhat faster on
4936 large queues. It won't affect queue runners, except when queue_run_in_order
4939 51. Installed eximstats 1.31 in the distribution.
4941 52. Added support for SRV lookups to the dnslookup router.
4943 53. If an ACL referred to $message_body or $message_body_end, the value was not
4944 reset for any messages that followed in the same SMTP session.
4946 54. The store-handling optimization for building very long strings was not
4947 differentiating between the different store pools. I don't think this
4948 actually made any difference in practice, but I've tidied it.
4950 55. While running the routers to verify a sender address, $sender_address
4951 was still set to the sender address. This is wrong, because when routing to
4952 send a bounce to the sender, it would be empty. Therefore, I have changed
4953 it so that, while verifying a sender address, $sender_address is set to <>.
4954 (There is no change to what happens when verifying a recipient address.)
4956 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
4957 target A or AAAA records (if not already returned) without resetting the
4958 qualify_single or search_parents options of the DNS resolver. These are
4959 inappropriate in this case because the targets of MX and SRV records must
4960 be FQDNs. A broken DNS record could cause trouble if it happened to have a
4961 target that, when qualified, matched something in the local domain. These
4962 two options are now turned off when doing these lookups.
4964 57. It seems that at least some releases of Reiserfs (which does not have the
4965 concept of a fixed number of inodes) returns zero and not -1 for the
4966 number of available inodes. This interacted badly with check_spool_inodes,
4967 which assumed that -1 was the "no such thing" setting. What I have done is
4968 to check that the total number of inodes is greater than zero before doing
4969 the test of how many are available.
4971 58. When a "warn" ACL statement has a log_message modifier, the message is
4972 remembered, and not repeated. This is to avoid a lot of repetition when a
4973 message has many recipients that cause the same warning to be written.
4974 Howewer, Exim was preserving the list of already written lines for an
4975 entire SMTP session, which doesn't seem right. The memory is now reset if a
4976 new message is started.
4978 59. The "rewrite" debugging flag was not showing the result of rewriting in the
4979 debugging output unless log_rewrite was also set.
4981 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
4982 of (int)(handle) when we know that handle contains (void *)(-1).
4984 61. The Exim daemon panic-logs an error return when it closes the incoming
4985 connection. However "connection reset by peer" seems to be common, and
4986 isn't really an error worthy of noting specially, so that particular error
4989 62. When Exim is trying to find all the local interfaces, it used to panic and
4990 die if the ioctl to get the interface flags failed. However, it seems that
4991 on at least one OS (Solaris 9) it is possible to have an interface that is
4992 included in the list of interfaces, but for which you get a failure error
4993 for this call. This happens when the interface is not "plumbed" into a
4994 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
4995 failure of the "get flags" call assumes that the interface is down.
4997 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
4998 makes life easier for people who are doing arithmetic on fields extracted
4999 from dates, where you often get leading zeros that should not be
5000 interpreted as octal.
5002 64. Added qualify_domain to the redirect router, to override the global
5005 65. If a pathologically long header line contained very many addresses (the
5006 report of this problem mentioned 10 000) and each of them was rewritten,
5007 Exim could use up a very large amount of memory. (It kept on making new
5008 copies of the header line as it rewrote, and never released the old ones.)
5009 At the expense of a bit more processing, the header rewriting function has
5010 been changed so that it no longer eats memory in this way.
5012 66. The generation of the Received: header has been moved from the time that a
5013 message starts to be received, to the time that it finishes. The timestamp
5014 in the Received: header should now be very close to that of the <= log
5015 line. There are two side-effects of this change:
5017 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5018 logged header lines no longer include the local Received: line, because
5019 it has not yet been created. The same applies to a copy of the message
5020 that is returned to a non-SMTP sender when a message is rejected.
5022 (b) When a filter file is tested using -bf, no additional Received: header
5023 is added to the test message. After some thought, I decided that this
5026 This change does not affect the value of $received_for. It is still set
5027 after address rewriting, but before local_scan() is called.
5029 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5031 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5032 gave an unhelpful panic error message, and a defer error. I have managed to
5033 change this behaviour so that it now rejects any supplied certificate,
5034 which seems right, as the list of acceptable certificates is empty.
5036 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5037 gave an unhelpful defer error. I have not managed to make this reject any
5038 supplied certificates, but the error message it gives is "no certificate
5039 supplied", which is not helpful.
5041 70. exigrep's output now also includes lines that are not associated with any
5042 message, but which match the given pattern. Implemented by a patch from
5043 Martin Sluka, which also tidied up the Perl a bit.
5045 71. Recipient callout verification, like sender verification, was using <> in
5046 the MAIL FROM command. This isn't really the right thing, since the actual
5047 sender may affect whether the remote host accepts the recipient or not. I
5048 have changed it to use the actual sender in the callout; this means that
5049 the cache record is now keyed on a recipient/sender pair, not just the
5050 recipient address. There doesn't seem to be a real danger of callout loops,
5051 since a callout by the remote host to check the sender would use <>.
5052 [SEE ABOVE: changed after hitting problems.]
5054 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5055 temporary errors. However, in the case of such a code being given after
5056 the end of a data transmission (i.e. after ".") Exim was failing to write
5057 a retry record for the message. (Yes, there was some broken host that was
5058 actually sending 8xx at this point.)
5060 73. An unknown lookup type in a host list could cause Exim to panic-die when
5061 the list was checked. (An example that provoked this was putting <; in the
5062 middle of a list instead of at the start.) If this happened during a DATA
5063 ACL check, a -D file could be left lying around. This kind of configuration
5064 error no longer causes Exim to die; instead it causes a defer errror. The
5065 incident is still logged to the main and panic logs.
5067 74. Buglet left over from Exim 3 conversion. The message "too many messages
5068 in one connection" was written to the rejectlog but not the mainlog, except
5069 when address rewriting (yes!) was being logged.
5071 75. Added write_rejectlog option.
5073 76. When a system filter was run not as root (that is, when system_filter_user
5074 was set), the values of the $n variables were not being returned to the
5075 main process; thus, they were not subsequently available in the $sn
5078 77. Added +return_path_on_delivery log selector.
5080 78. A connection timeout was being treated differently from recipients deferred
5081 when testing hosts_max_try with a message that was older than the host's
5082 retry timeout. (The host should not be counted, thus allowing all hosts to
5083 be tried at least once before bouncing.) This may have been the cause of an
5084 occasionally reported bug whereby a message would remain on the queue
5085 longer than the retry timeout, but would be bounced if a delivery was
5086 forced. I say "may" because I never totally pinned down the problem;
5087 setting up timeout/retry tests is difficult. See also the next item.
5089 79. The ultimate address timeout was not being applied to errors that involved
5090 a combination of host plus message (for example, a timeout on a MAIL
5091 command). When an address resolved to a number of possible hosts, and they
5092 were not all tried for each delivery (e.g. because of hosts_max_try), a
5093 message could remain on the queue longer than the retry timeout.
5095 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5098 81. Fixed an obscure SMTP outgoing bug which required at least the following
5099 conditions: (a) there was another message waiting for the same server;
5100 (b) the server returned 5xx to all RCPT commands in the first message so
5101 that the message was not completed; (c) the server dropped the connection
5102 or gave a negative response to the RSET that Exim sends to abort the
5103 transaction. The observed case was a dropped connection after DATA that had
5104 been sent in pipelining mode. That is, the server had advertised PIPELINING
5105 but was not implementing it correctly. The effect of the bug was incorrect
5106 behaviour, such as trying another host, and this could lead to a crash.
5112 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5113 and daemon.c were passed as pointers to ints; they should have been
5114 pointers to socklen_t variables (which are typically unsigned ints).
5116 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5119 3. Fixed a really odd bug that affected only the testing scheme; patching a
5120 certain fixed string in the binary changed the value of another string that
5121 happened to be identical to the end of the original first string.
5123 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5124 name", it returns that address as the IP address. On some operating
5125 systems (e.g. Solaris), it also passes back the IP address string as the
5126 "host name". However, on others (e.g. Linux), it passes back an empty
5127 string. Exim wasn't checking for this, and was changing the host name to an
5128 empty string, assuming it had been canonicized.
5130 5. Although rare, it is permitted to have more than one PTR record for a given
5131 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5132 all the names associated with an address, because they do in Solaris.
5133 However, it seems that they do not in Linux for data that comes from the
5134 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5135 I found this out when I moved to a new Linux workstation and tried to run
5136 the Exim test suite.
5138 To get round this problem I have changed the code so that it now does its
5139 own call to the DNS to look up PTR records when searching for a host name.
5140 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5141 addresses that are only in /etc/hosts are still found.
5143 This behaviour is, however, controlled by an option called host_lookup_
5144 order, which defaults to "bydns:byaddr". If people want to use the other
5145 order, or indeed, just use one or the other means of lookup, they can
5146 specify it in this variable.
5148 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5149 some operating systems, this comes back from gethostbyaddr() as an empty
5150 string, and this is what Exim used to test for. However, it seems that in
5151 other systems, "." is yielded. Exim now tests for this case too.
5153 7. The values of check_spool_space and check_log_space are now held internally
5154 as a number of kilobytes instead of an absolute number of bytes. If a
5155 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5156 kilobyte. This means that much larger values can be stored.
5158 8. Exim monitor: an attempt to get the action menu when not actually pointing
5159 at a message produces an empty menu entitled "No message selected". This
5160 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5161 no entries in it ("Shell widget menu has zero width and/or height"). So I
5162 have added a single, blank menu entry in this case.
5164 9. Added ${quote_local_part.
5166 10. MIME decoding is now applied to the contents of Subject: header lines when
5169 11. Now that a reference to $sender_host_address automatically causes a reverse
5170 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5171 host lookup before query-style lookups in lists that might use this
5172 variable. This has therefore been abolished, and the "net-" prefix is no
5173 longer necessary for query-style lookups.
5175 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5176 have been a typo for LFLAGS, so it has been changed.
5178 13. The install script calls Exim with "-C /dev/null" in order to find the
5179 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5180 to be output. Howeve, since Exim outputs its version number before the
5181 error, it didn't break the script. It just looked ugly. I fixed this by
5182 always allowing "-C /dev/null" if the caller is root.
5184 14. Ignore overlarge ACL variable number when reading spool file - insurance
5185 against a later release with more variables having written the file.
5187 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5188 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5189 this, as well as the form without the "IPv6" on the front.
5191 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5192 OS/Makefile-Darwin file.
5194 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5195 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5197 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5198 variable. However, some OS use ints for inodes. I've added cast to long int
5199 to get rid of the compiler warning.
5201 19. I had forgotten to lock out "/../" in configuration file names when
5202 ALT_CONFIG_PREFIX was set.
5204 20. Routers used for verification do not need to specify transports. However,
5205 if such a router generated a host list, and callout was configured, Exim
5206 crashed, because it could not find a port number from the (non-existent)
5207 transport. It now assumes port 25 in this circumstance.
5209 21. Added the -t option to exigrep.
5211 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5212 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5213 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5214 them is set, LOOKUP_LSEARCH is forced.
5216 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5218 24. Added sender and host information to the "rejected by local_scan()" log
5219 line; previously there was no indication of these.
5221 25. Added .include_if_exists.
5223 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5224 Linux. It turns out that not all file systems support this. Apparently some
5225 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5226 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5227 not supported on the file descriptor, is now ignored when Exim is trying to
5228 sync a directory. This applies only to Linux.
5230 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5232 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5233 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5236 29. Added support for maildirsize files from supplied patch (modified a bit).
5238 30. The use of :fail: followed by an empty string could lead Exim to respond to
5239 sender verification failures with (e.g.):
5241 550 Verification failed for <xxx>
5242 550 Sender verify failed
5244 where the first response line was missing the '-' that indicates it is not
5245 the final line of the response.
5247 31. The loop for finding the name of the user that called Exim had a hardwired
5248 limit of 10; it now uses the value of finduser_retries, which is used for
5249 all other user lookups.
5251 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5254 33. Exim was neglecting to zero errno before one call of strtol() when
5255 expanding a string and expecting an integer value. On some systems this
5256 resulted in spurious "integer overflow" errors. Also, it was casting the
5257 result into an int without checking.
5259 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5260 did not work. The code looks as if it has *never* worked, though it appears
5261 to have been documented since at least releast 1.62. I have made it work.
5263 35. The "timeout_DNS" error in retry rules, also documented since at least
5264 1.62, also never worked. As it isn't clear exactly what this means, and
5265 clearly it isn't a major issue, I have abolished the feature by treating it
5266 as "timeout", and writing a warning to the main and panic logs.
5268 36. The display of retry rules for -brt wasn't always showing the error code
5271 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5272 timeout_connect_A, timeout_connect_MX.
5274 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5275 to the empty sender.
5277 39. The daemon was not analysing the content of -oX till after it had closed
5278 stderr and disconnected from the controlling terminal. This meant that any
5279 syntax errors were only noted on the panic log, and the return code from
5280 the command was 0. By re-arranging the code a little, I've made the
5281 decoding happen first, so such errors now appear on stderr, and the return
5282 code is 1. However, the actual setting up of the sockets still happens in
5283 the disconnected process, so errors there are still only recorded on the
5286 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5287 connections (as happens on some IP stacks) was logged at start up time as
5288 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5289 it from "IPv6 and IPv4", which means that two separate sockets are being
5292 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5293 says whether AF_INET or AF_INET6 was passed as an argument.
5295 42. Exiwhat output was messed up when time zones were included in log
5298 43. Exiwhat now gives more information about the daemon's listening ports,
5299 and whether -tls-on-connect was used.
5301 44. The "port" option of the smtp transport is now expanded.
5303 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5304 silently ignored. Now an error message is written to the main and panic
5307 46. There's a new ACL modifier called "logwrite" which writes to a log file
5308 as soon as it is encountered.
5310 47. Added $local_user_uid and $local_user_gid at routing time.
5312 48. Exim crashed when trying to verify a sender address that was being
5315 49. Exim was recognizing only a space character after ".include". It now also
5316 recognizes a tab character.
5318 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5319 extracting the relevant information from the specification. The man page no
5320 longer contains scrambled data for the -d option, and I've added a section
5321 at the front about calling Exim under different names.
5323 51. Added "extra_headers" argument to the "mail" command in filter files.
5325 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5328 53. Installed eximstats 1.29.
5330 54. Added transport_filter_timeout as a generic transport option.
5332 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5333 Cc: header lines. This was required by RFC 822, but it not required by RFC
5336 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5337 incoming messages that did not have them. Now it does so only if the
5338 message originates locally, that is, if there is no associated remote host
5339 address. When Resent- header lines are present, this applies to the Resent-
5340 lines rather than the non-Resent- lines.
5342 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5343 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5345 58. Messages for configuration errors now include the name of the main
5346 configuration file - useful now that there may be more than one file in a
5347 list (.included file names were always shown).
5349 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5350 for those rare installations that do not start the daemon as root or run it
5351 setuid root. I've cut out the call to initgroups() if the daemon is not
5354 60. The Exim user and group can now be bound into the binary as text strings
5355 that are looked up at the start of Exim's processing.
5357 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5359 62. Added $mailstore_basename variable.
5361 63. Installed patch to sieve.c from Michael Haardt.
5363 64. When Exim failed to open the panic log after failing to open the main log,
5364 the original message it was trying to log was written to stderr and debug
5365 output, but if they were not available (the usual case in production), it
5366 was lost. Now it is written to syslog before the two lines that record the
5367 failures to open the logs.
5369 65. Users' Exim filters run in subprocesses under the user's uid. It is
5370 possible for a "deliver" command or an alias in a "personal" command to
5371 provoke an address rewrite. If logging of address rewriting is configured,
5372 this fails because the process is not running as root or exim. There may be
5373 a better way of dealing with this, but for the moment (because 4.30 needs
5374 to be released), I have disabled address rewrite logging when running a
5375 filter in a non-root, non-exim process.
5381 1. The buildconfig auxiliary program wasn't quoting the value set for
5382 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5383 not defined. This bug was masked in 4.22 by the effect that was fixed in
5386 2. Some messages that were rejected after a message id was allocated were
5387 shown as "incomplete" by exigrep. It no longer does this for messages that
5388 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5390 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5391 have allow_domain_literals set, the ID did not get logged in the <= line.
5392 Domain literals are now always recognized in Message-ID: header lines.
5394 4. The first argument for a ${extract expansion item is the key name or field
5395 number. Leading and trailing spaces in this item were not being ignored,
5396 causing some misleading effects.
5398 5. When deliver_drop_privilege was set, single queue runner processes started
5399 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5400 same command in the process it spins off) were not dropping privilege.
5402 6. When the daemon running as "exim" started a queue runner, it always
5403 re-executed Exim in the spun-off process. This is a waste of effort when
5404 deliver_drop_privilege is set. The new process now just calls the
5405 queue-runner function directly.
5411 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5414 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5415 ignored. Though the use of -oP was forcing the writing of a pid file, it
5416 was always written to the default place.
5418 3. If the message "no IP address found for host xxxx" is generated during
5419 incoming verification, it is now followed by identification of the incoming
5420 connection (so you can more easily find what provoked it).
5422 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5424 5. Added some features to "harden" Exim a bit more against certain attacks:
5426 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5427 be put in Local/Makefile. This is like the never_users runtime option,
5428 but it cannot be overridden. The default setting is "root".
5430 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5431 prefix string with which any file named in a -C command line option
5434 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5435 is retained for -C and -D only if the caller of Exim is root. Without
5436 it, the exim user may also use -C and -D and retain privilege.
5438 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5439 command line option is disabled.
5441 6. Macro names set by the -D option must start with an upper case letter, just
5442 like macro names defined in the configuration file.
5444 7. Added "dereference=" facility to LDAP.
5446 8. Two instances of the typo "uknown" in the source files are fixed.
5448 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5449 the Configure-Makefile script screwed up while processing it.
5451 10. Incorporated PCRE 4.4.
5453 11. The SMTP synchronization check was not operating right at the start of an
5454 SMTP session. For example, it could not catch a HELO sent before the client
5455 waited for the greeting. There is now a check for outstanding input at the
5456 point when the greeting is written. Because of the duplex, asynchronous
5457 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5458 way, but not yet received, when the check is performed.
5460 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5461 on TCP/IP sockets, because this apparently causes some broken clients to
5464 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5465 unchanged) from the Cygwin maintainer.
5467 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5468 was supported instead of testing for mbx. Effectively a typo.
5470 15. The spa authenticator server code was not checking that the input it
5471 received was valid base64.
5473 16. The debug output line for the "set" modifier in ACLs was not showing the
5474 name of the variable that was being set.
5476 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5478 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5479 lookup on its own. Something else was needed to trigger the lookup. For
5480 example, a match in host_lookup or the need for a host name in a host list.
5481 Now, if $sender_host_name is referenced and the host name has not yet been
5482 looked up, a lookup is performed. If the lookup fails, the variable remains
5483 empty, and $host_lookup_failed is set to "1".
5485 19. Added "eqi" as a case-independent comparison operator.
5487 20. The saslauthd authentication condition could segfault if neither service
5488 nor realm was specified.
5490 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5491 error message that was logged was misleading, and incoming SMTP
5492 connections were dropped. The message is now more accurate, and temporary
5493 errors are given to SMTP connections.
5495 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5496 (except RSET) with a 421 error, until QUIT is received. However, it was
5497 failing to send a response to QUIT.
5499 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5500 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5501 ACL is now run after the helo_try_verify_hosts code.
5503 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5504 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5505 case-independent, so other case variants are also recognized). Apparently
5506 some systems use these upper case variants.
5508 25. If more than two messages were waiting for the same host, and a transport
5509 filter was specified for the transport, Exim sent two messages over the
5510 same TCP/IP connection, and then failed with "socket operation on non-
5511 socket" when it tried to send the third.
5513 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5515 27. The extern definition of crypt16() in expand.c was not being excluded when
5516 the OS had its own crypt16() function.
5518 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5519 as a preferred synonym for return_size_limit, both as an option and as an
5522 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5524 30. Changed the default configuration ACL to relax the local part checking rule
5525 for addresses that are not in any local domains. For these addresses,
5526 slashes and pipe symbols are allowed within local parts, but the sequence
5527 /../ is explicitly forbidden.
5529 31. SPA server authentication was not clearing the challenge buffer before
5532 32. log_message in a "warn" ACL statement was writing to the reject log as
5533 well as to the main log, which contradicts the documentation and doesn't
5534 seem right (because no rejection is happening). So I have stopped it.
5536 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5537 However, I am unable to do any testing of this.
5539 34. Fixed an infelicity in the appendfile transport. When checking directories
5540 for a mailbox, to see if any needed to be created, it was accidentally
5541 using path names with one or more superfluous leading slashes; tracing
5542 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5544 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5545 contents are added to the log line that is written for every discarded
5546 recipient. (Previously a log_message setting was ignored.)
5548 36. The ${quote: operator now quotes the string if it is empty.
5550 37. The install script runs exim in order to find its version number. If for
5551 some reason other than non-existence or emptiness, which it checks, it
5552 could not run './exim', it was installing it with an empty version number,
5553 i.e. as "exim-". This error state is now caught, and the installation is
5556 38. An argument was missing from the function that creates an error message
5557 when Exim fails to connect to the socket for saslauthd authentication.
5558 This could cause Exim to crash, or give a corrupted message.
5560 39. Added isip, isip4, and isip6 to ${if conditions.
5562 40. The ACL variables $acl_xx are now saved with the message, and can be
5563 accessed later in routers, transports, and filters.
5565 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
5566 strings in the file are not string-expanded.
5568 42. If a MAIL command specified a SIZE value that was too large to fit into an
5569 int variable, the check against message_size_limit failed. Such values are
5570 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
5571 day this will have to be increased, but I don't think I want to be around
5572 when emails are that large.
5579 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
5580 iconv() is not standard in FreeBSD.
5582 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
5583 IPv6 enabled. The observed symptom was a segmentation fault on return from
5584 the function os_common_find_running_interfaces() in src/os.c.
5586 3. In the check_special_case() function in daemon.c I had used "errno" as an
5587 argument name, which causes warnings on some systems. This was basically a
5588 typo, since it was named "eno" in the comments!
5590 4. The code that waits for the clock to tick (at a resolution of some fraction
5591 of a second) so as to ensure message-id uniqueness was always waiting for
5592 at least one whole tick, when it could have waited for less. [This is
5593 almost certainly not relevant at current processor speeds, where it is
5594 unlikely to ever wait at all. But we try to future-proof.]
5596 5. The function that sleeps for a time interval that includes fractions of a
5597 second contained a race. It did not block SIGALRM between setting the
5598 timer, and suspending (a couple of lines later). If the interval was short
5599 and the sigsuspend() was delayed until after it had expired, the suspension
5600 never ended. On busy systems this could lead to processes getting stuck for
5603 6. Some uncommon configurations may cause a lookup to happen in a queue runner
5604 process, before it forks any delivery processes. The open lookup caching
5605 mechanism meant that the open file or database connection was passed into
5606 the delivery process. The problem was that delivery processes always tidy
5607 up cached lookup data. This could cause a problem for the next delivery
5608 process started by the queue runner, because the external queue runner
5609 process does not know about the closure. So the next delivery process
5610 still has data in the lookup cache. In the case of a file lookup, there was
5611 no problem because closing a file descriptor in a subprocess doesn't affect
5612 the parent. However, if the lookup was caching a connection to a database,
5613 the connection was closed, and the second delivery process was likely to
5614 see errors such as "PGSQL: query failed: server closed the connection
5615 unexpectedly". The problem has been fixed by closing all cached lookups
5616 in a queue runner before running a delivery process.
5618 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
5619 seem to have the "const" qualifier which it has on other OS. I've
5622 8. Change 4.21/2 was too strict. It is only if there are two authenticators
5623 *of the same type* (client or server) with the same public name that an
5624 error should be diagnosed.
5626 9. When Exim looked up a host name for an IP address, but failed to find the
5627 original IP address when looking up the host name (a safety check), it
5628 output the message "<ip address> does not match any IP for NULL", which was
5629 confusing, to say the least. The bug was that the host name should have
5630 appeared instead of "NULL".
5632 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
5633 user that is built into the binary, and the -C or -D options is used, root
5634 privilege is dropped before the configuration file is read. In addition,
5635 logging is switched to stderr instead of the normal log files. If the
5636 configuration then re-defines the Exim user, the unprivileged environment
5637 is probably not what is expected, so Exim logs a panic warning message (but
5640 However, if deliver_drop_privilege is set, the unprivileged state may well
5641 be exactly what is intended, so the warning has been cut out in that case,
5642 and Exim is allowed to try to write to its normal log files.
5648 1. smtp_return_error_details was not giving details for temporary sender
5649 or receiver verification errors.
5651 2. Diagnose a configuration error if two authenticators have the same public
5654 3. Exim used not to create the message log file for a message until the first
5655 delivery attempt. This could be confusing when incoming messages were held
5656 for policy or load reasons. The message log file is now created at the time
5657 the message is received, and an initial "Received" line is written to it.
5659 4. The automatically generated man page for command line options had a minor
5660 bug that caused no ill effects; however, a more serious problem was that
5661 the procedure for building the man page automatically didn't always
5662 operate. Consequently, release 4.20 contains an out-of-date version. This
5663 shouldn't happen again.
5665 5. When building Exim with embedded Perl support, the script that builds the
5666 Makefile was calling 'perl' to find its compile-time parameters, ignoring
5667 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
5669 6. The freeze_tell option was not being used for messages that were frozen on
5670 arrival, either by an ACL or by local_scan().
5672 7. Added the smtp_incomplete_transaction log selector.
5674 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
5675 was accepting AUTH without a new EHLO.
5677 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
5678 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
5681 10. Logging of TCP/IP connections (when configured) now happens in the main
5682 daemon process instead of the child process, so that the TCP/IP connection
5683 count is more accurate (but it can never be perfect).
5685 11. The use of "drop" in a nested ACL was not being handled correctly in the
5686 outer ACL. Now, if condition failure induced by the nested "drop" causes
5687 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
5688 or "require"), the connection is dropped.
5690 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
5691 that yield "discard" can now be used with an "accept" or a "discard" verb,
5692 but an error is generated for any others (because I can't see a useful way
5693 to define what should happen).
5695 13. When an ACL is read dynamically from a file (or anywhere else), the lines
5696 are now processed in the same way as lines in the Exim configuration file.
5697 In particular, continuation lines are supported.
5699 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
5701 15. Added -ti meaning -t -i.
5703 16. Check for letters, digits, hyphens, and dots in the names of dnslist
5704 domains, and warn by logging if others are found.
5706 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
5707 returned from GIFCONF when Exim is trying to find the list of interfaces on
5708 a host. The code in os.c has been modified to copy each ifreq to an aligned
5709 structure in all cases.
5711 Also, in some cases, the returned ifreq's were being copied to a 'struct
5712 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
5713 means the last couple of bytes of an IPv6 address could be chopped if the
5714 ifreq contained only a normal sockaddr (14 bytes storage).
5716 18. Named domain lists were not supported in the hosts_treat_as_local option.
5717 An entry such as +xxxx was not recognized, and was treated as a literal
5720 19. Ensure that header lines added by a DATA ACL are included in the reject log
5721 if the ACL subsequently rejects the message.
5723 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
5724 MD5 (which is deprecated).
5726 21. When testing a filter file using -bf, Exim was writing a message when it
5727 took the sender from a "From " line in the message, but it was not doing so
5728 when it took $return_path from a Return-Path: header line. It now does.
5730 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
5731 with a valid header line field name (a series of printing characters
5732 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
5734 23. Changed "disc" in the source to "disk" to conform to the documentation and
5735 the book and for uniformity.
5737 24. Ignore Sendmail's -Ooption=value command line item.
5739 25. When execve() failed while trying to run a command in a pipe transport,
5740 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
5741 could be confused with a return value of 69 from the command itself. This
5742 has been changed to 127, the value the shell returns if it is asked to run
5743 a non-existent command. The wording for the related log line suggests a
5744 non-existent command as the problem.
5746 26. If received_header_text expands to an empty string, do not add a Received:
5747 header line to the message. (Well, it adds a token one on the spool, but
5748 marks it "old" so that it doesn't get used or transmitted.)
5750 27. Installed eximstats 1.28 (addition of -nt option).
5752 28. There was no check for failure on the call to getsockname() in the daemon
5753 code. This can fail if there is a shortage of resources on the system, with
5754 ENOMEM, for example. A temporary error is now given on failure.
5756 29. Contrary to the C standard, it seems that in some environments, the
5757 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
5758 program. Exim now does this explicitly; it affects the formatting of
5759 timestamps using strftime().
5761 30. If exiqsumm was given junk data, it threw up some uninitialized variable
5762 complaints. I've now initialized all the variables, to avoid this.
5764 32. Header lines added by a system filter were not being "seen" during
5765 transport-time rewrites.
5767 33. The info_callback() function passed to OpenSSL is set up with type void
5768 (*)(SSL *, int, int), as described somewhere. However, when calling the
5769 function (actually a macro) that sets it up, the type void(*)() is
5770 expected. I've put in a cast to prevent warnings from picky compilers.
5772 34. If a DNS black list lookup found a CNAME record, but there were no A
5773 records associated with the domain it pointed at, Exim crashed.
5775 35. If a DNS black list lookup returned more than one A record, Exim ignored
5776 all but the first. It now scans all returned addresses if a particular IP
5777 value is being sought. In this situation, the contents of the
5778 $dnslist_value variable are a list of all the addresses, separated by a
5781 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
5782 to accept a host name and all its aliases if the forward lookup for any of
5783 them yielded the IP address of the incoming connection. Now it accepts only
5784 those names whose forward lookup yields the correct IP address. Any other
5785 names are discarded. This closes a loophole whereby a rogue DNS
5786 administrator could create reverse DNS records to break through a
5787 wildcarded host restriction in an ACL.
5789 37. If a user filter or a system filter that ran in a subprocess used any of
5790 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
5791 the wrong values were passed to the pipe command ($thisaddress had the
5792 value of $0, $0 had the value of $1, etc). This bug was introduced by
5793 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
5795 38. Improved the line breaking for long SMTP error messages from ACLs.
5796 Previously, if there was no break point between 40 and 75 characters, Exim
5797 left the rest of the message alone. Two changes have been made: (a) I've
5798 reduced the minimum length to 35 characters; (b) if it can't find a break
5799 point between 35 and 75 characters, it looks ahead and uses the first one
5800 that it finds. This may give the occasional overlong line, but at least the
5801 remaining text gets split now.
5803 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
5804 file descriptors might be low, and that setting 1000 would always raise it.
5805 It turns out that in some environments, the limit is already over 1000 and
5806 that lowering it causes trouble. So now Exim takes care not to decrease it.
5808 40. When delivering a message, the value of $return_path is set to $sender_
5809 address at the start of routing (routers may change the value). By an
5810 oversight, this default was not being set up when an address was tested by
5811 -bt or -bv, which affected the outcome if any router or filter referred to
5814 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
5815 log only when "message" or "log_message" are set. However, if one of the
5816 conditions was an address verification, or a call to a nested ACL, the
5817 messages generated by the underlying test were being passed through. This
5818 no longer happens. The underlying message is available in $acl_verify_
5819 message for both "message" and "log_message" expansions, so it can be
5820 passed through if needed.
5822 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
5823 new expansion $bh_ to give the encoded byte string without charset
5824 translation. Translation happens only if iconv() is available; HAVE_ICONV
5825 indicates this at build time. HEADERS_CHARSET gives the charset to
5826 translate to; headers_charset can change it in the configuration, and
5827 "headers charset" can change it in an individual filter file.
5829 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
5830 that creates RFC 2047 encoded "words" labels them as that charset instead
5831 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
5832 expansion operator; (ii) when Exim creates a From: line for a local
5833 message; (iii) when a header line is rewritten to include a "phrase" part.
5835 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
5836 buggy, causing it to skip the first lines of messages whose message ID
5837 ended in 'D'. This would not have bitten before Exim release 4.14, because
5838 message IDs were unlikely to end in 'D' before then. The effect was to have
5839 incorrect size information for certain domains.
5841 45. #include "config.h" was missing at the start of the crypt16.c module. This
5842 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
5845 46. If there was a timeout during a "random" callout check, Exim treated it as
5846 a failure of the random address, and carried on sending RSET and the real
5847 address. If the delay was just some slowness somewhere, the response to the
5848 original RCPT would be taken as a response to RSET and so on, causing
5849 mayhem of various kinds.
5851 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
5852 when I implemented it. It didn't allow for the fact that some option values
5853 may legitimatetly be negative (e.g. size_addition), and it didn't even do
5854 the right test for positive values.
5856 48. Domain names in DNS records are case-independent. Exim always looks them up
5857 in lower case. Some resolvers return domain names in exactly the case they
5858 appear in the zone file, that is, they may contain uppercase letters. Not
5859 all resolvers do this - some return always lower case. Exim was treating a
5860 change of case by a resolver as a change of domain, similar to a widening
5861 of a domain abbreviation. This triggered its re-routing code and so it was
5862 trying to route what was effectively the same domain again. This normally
5863 caused routing to fail (because the router wouldn't handle the domain
5864 twice). Now Exim checks for this case specially, and just changes the
5865 casing of the domain that it ultimately uses when it transmits the message
5868 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
5871 50. If a filter generated a file delivery with a non-absolute name (possible if
5872 no home directory exists for the router), the forbid_file option was not
5875 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
5876 the existing equality matching.
5878 52. Exim was using ints instead of ino_t variables in some places where it was
5879 dealing with inode numbers.
5881 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
5882 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
5883 called TMPDIR, and if it finds it is different, it changes its value.
5885 54. The smtp_printf() function is now made available to local_scan() so
5886 additional output lines can be written before returning. There is also an
5887 smtp_fflush() function to enable the detection of a dropped connection.
5888 The variables smtp_input and smtp_batched_input are exported to
5891 55. Changed the default runtime configuration: the message "Unknown user"
5892 has been removed from the ACL, and instead placed on the localuser router,
5893 using the cannot_route_message feature. This means that any verification
5894 failures that generate their own messages won't get overridden. Similarly,
5895 the "Unrouteable address" message that was in the ACL for unverifiable
5896 relay addresses has also been removed.
5898 56. Added hosts_avoid_esmtp to the smtp transport.
5900 57. The exicyclog script was not checking for the esoteric option
5901 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
5902 will work only if exicyclog is run under the appropriate euid.
5904 58. Following a discussion on the list, the rules by which Exim recognises line
5905 endings on incoming messages have been changed. The -dropcr and drop_cr
5906 options are now no-ops, retained only for backwards compatibility. The
5907 following line terminators are recognized: LF CRLF CR. However, special
5908 processing applies to CR:
5910 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
5911 nor a local message in the state where . is a terminator.
5913 (ii) If a bare CR is encountered in a header line, an extra space is added
5914 after the line terminator so as not to end the header. The reasoning
5915 behind this is that bare CRs in header lines are most likely either
5916 to be mistakes, or people trying to play silly games.
5918 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
5919 was being incorrectly given as 18 bytes larger than it should have been.
5920 This is a VOB (very old bug).
5922 60. This may never have affected anything current, but just in case it has:
5923 When the local host is found other than at the start of a list of hosts,
5924 the local host, those with the same MX, and any that follow, are discarded.
5925 When the list in question was part of a longer list of hosts, the following
5926 hosts (not currently being processed) were also being discarded. This no
5927 longer happens. I'm not sure if this situation could ever has previously
5930 61. Added the "/MX" feature to lists of hosts in the manualroute and query
5933 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
5934 header. This is something that is recommended in a new Internet Draft, and
5935 is something that is documented as being done by Sendmail. There are two
5936 possible values. For messages generated by the autoreply transport, Exim
5939 Auto-Submitted: auto-replied
5941 whereas for all other generated messages (e.g. bounces) it adds
5943 Auto-Submitted: auto-generated
5945 63. The "personal" condition in filters now includes a test for the
5946 Auto-Submitted: header. If it contains the string "auto-" the message it
5947 not considered personal.
5949 64. Added rcpt_include_affixes as a generic transport option.
5951 65. Added queue_only_override (default true).
5953 66. Added the syslog_duplication option.
5955 67. If what should have been the first header line of a message consisted of
5956 a space followed by a colon, Exim was mis-interpreting it as a header line.
5957 It isn't of course - it is syntactically invalid and should therefore be
5958 treated as the start of the message body. The misbehaviour could have
5959 caused a number of strange effects, including loss of data in subsequent
5960 header lines, and spool format errors.
5962 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
5963 client host had authenticated. This control can now be exercised by an ACL
5964 for more flexibility.
5966 69. By default, callouts do not happen when testing with -bh. There is now a
5967 variant, -bhc, which does actually run the callout code, including
5968 consulting and updating the callout cache.
5970 70. Added support for saslauthd authentication, courtesy of Alexander
5973 71. If statvfs() failed on the spool or log directories while checking their
5974 size for availability, Exim confusingly gave the error "space shortage".
5975 Furthermore, in debugging mode it crashed with a floating point exception.
5976 These checks are done if check_{spool,log}_{space,inodes} are set, and when
5977 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
5978 serious problem, Exim now writes to the main and panic logs when this
5979 happens, with details of the failure. It then refuses to accept the
5980 incoming message, giving the message "spool directory problem" or "log
5981 directory problem" with a 421 code for SMTP messages.
5983 72. When Exim is about to re-exec itself, it ensures that the file descriptors
5984 0, 1, and 2 exist, because some OS complain for execs without them (see
5985 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
5986 descriptors. However, the code omitted to check that the open succeeded,
5987 causing mysterious errors if for some reason the permissions on /dev/null
5988 got screwed. Now Exim writes a message to the main and panic logs, and
5989 bombs out if it can't open /dev/null.
5991 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
5992 interact so that it is all more flexible. It is supposed to remain
5993 backwards compatible. Also added extra_local_interfaces.
5995 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
5996 to bomb out with an assertion failure - to the client this appears as a
5997 connection drop. This problem occurs in the part of the code that was taken
5998 from the Samba project. Fortunately, the assertion is in a very simple
5999 function, so I have fixed this by reproducing the function inline in the
6000 one place where it is called, and arranging for authentication to fail
6001 instead of killing the process with assert().
6003 75. The SPA client code was not working when the server requested OEM rather
6004 than Unicode encoding.
6006 76. Added code to make require_files with a specific uid setting more usable in
6007 the case where statting the file as root fails - usually a non-root-mounted
6008 NFS file system. When this happens and the failure is EACCES, Exim now
6009 forks a subprocess and does the per-uid checking as the relevant uid.
6011 77. Added process_log_path.
6013 78. If log_file_path was not explicitly set, a setting of check_log_space or
6014 check_log_inodes was ignored.
6016 79. If a space check for the spool or log partitions fails, the incident is now
6017 logged. Of course, in the latter case the data may get lost...
6019 80. Added the %p formatting code to string_format() so that it can be used to
6020 print addresses in debug_print(). Adjusted all the address printing in the
6021 debugging in store.c to use %p rather than %d.
6023 81. There was a concern that a line of code in smtp_in.c could overflow a
6024 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6025 initially expressed, the concern was not well-founded, because trailing
6026 spaces are removed early. However, if the trailing spaces were followed by
6027 a NULL, they did not get removed, so the overflow was possible. Two fixes
6030 (a) I re-wrote the offending code in a cleaner fashion.
6031 (b) If an incoming SMTP command contains a NULL character, it is rejected
6034 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6035 runs initgroups(), so that if the Exim user is in any additional groups,
6036 they will be used during message reception.
6042 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob