1 Change log file for Exim from version 4.21
2 ------------------------------------------
3 This document describes *changes* to previous versions, that might
4 affect Exim's operation, with an unchanged configuration file. For new
5 options, and new features, see the NewStuff file next to this ChangeLog.
11 JH/01 Avoid costly startup code when not strictly needed. This reduces time
12 for some exim process initialisations. It does mean that the logging
13 of TLS configuration problems is only done for the daemon startup.
15 JH/02 Early-pipelining support code is now included unless disabled in Makefile.
17 JH/03 DKIM verification defaults no long accept sha1 hashes, to conform to
18 RFC 8301. They can still be enabled, using the dkim_verify_hashes main
21 JH/04 Support CHUNKING from an smtp transport using a transport_filter, when
22 DKIM signing is being done. Previously a transport_filter would always
23 disable CHUNKING, falling back to traditional DATA.
25 JH/05 Regard command-line receipients as tainted.
27 JH/06 Bug 340: Remove the daemon pid file on exit, whe due to SIGTERM.
33 JH/01 OpenSSL: With debug enabled output keying information sufficient, server
34 side, to decode a TLS 1.3 packet capture.
36 JH/02 OpenSSL: Suppress the sending of (stateful) TLS1.3 session tickets.
37 Previously the default library behaviour applied, sending two, each in
40 JH/03 Debug output for ACL now gives the config file name and line number for
43 JH/04 The default received_header_text now uses the RFC 8314 tls cipher clause.
45 JH/05 DKIM: ensure that dkim_domain elements are lowercased before use.
47 JH/06 Fix buggy handling of autoreply bounce_return_size_limit, and a possible
48 buffer overrun for (non-chunking) other transports.
50 JH/07 GnuTLS: Our use of late (post-handshake) certificate verification, under
51 TLS1.3, means that a server rejecting a client certificate is not visible
52 to the client until the first read of encrypted data (typically the
53 response to EHLO). Add detection for that case and treat it as a failed
54 TLS connection attempt, so that the normal retry-in-clear can work (if
57 JB/01 Bug 2375: fix expansions of 822 addresses having comments in local-part
58 and/or domain. Found and fixed by Jason Betts.
60 JH/08 Add hardening against SRV & TLSA lookups the hit CNAMEs (a nonvalid
61 configuration). If a CNAME target was not a wellformed name pattern, a
64 JH/09 Logging: Fix initial listening-on line for multiple ports for an IP when
65 the OS reports them interleaved with other addresses.
67 JH/10 OpenSSL: Fix aggregation of messages. Previously, when PIPELINING was
68 used both for input and for a verify callout, both encrypted, SMTP
69 responses being sent by the server could be lost. This resulted in
70 dropped connections and sometimes bounces generated by a peer sending
73 JH/11 Harden plaintext authenticator against a badly misconfigured client-send
74 string. Previously it was possible to cause undefined behaviour in a
75 library routine (usually a crash). Found by "zerons".
77 JH/12 Bug 2384: fix "-bP smtp_receive_timeout". Previously it returned no
80 JH/13 Bug 2386: Fix builds with Dane under LibreSSL 2.9.0 onward. Some old
81 API was removed, so update to use the newer ones.
83 JH/14 Bug 1891: Close the log file if receiving a non-smtp message, without
84 any timeout set, is taking a long time. Previously we would hang on to a
85 rotated logfile "forever" if the input was arriving with long gaps
86 (a previous attempt to fix addressed lack, for a long time, of initial
89 HS/01 Bug 2390: Use message_id for tempfile creation to avoid races in a
90 shared (NFS) environment. The length of the tempfile name is now
91 4 + 16 ("hdr.$message_exim_id") which might break on file
92 systems which restrict the file name length to lower values.
95 HS/02 Bug 2390: Use message_id for tempfile creation to avoid races in a
96 shared (NFS) environment.
98 HS/03 Bug 2392: exigrep does case sensitive *option* processing (as it
99 did for all versions <4.90). Notably -M, -m, --invert, -I may be
102 JH/15 Use unsigned when creating bitmasks in macros, to avoid build errors
103 on some platforms for bit 31.
105 JH/16 GnuTLS: rework ciphersuite strings under recent library versions. Thanks
106 to changes apparently associated with TLS1.3 handling some of the APIs
107 previously used were either nonfunctional or inappropriate. Strings
108 like TLS1.3:ECDHE_SECP256R1__RSA_PSS_RSAE_SHA256__AES_256_GCM__AEAD:256
109 and TLS1.2:ECDHE_SECP256R1__RSA_SHA256__AES_128_CBC__SHA256:128 replace
110 the previous TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256 .
111 This affects log line X= elements, the $tls_{in,out}_cipher variables,
112 and the use of specific cipher names in the encrypted= ACL condition.
114 JH/17 OpenSSL: the default openssl_options now disables ssl_v3.
116 JH/18 GnuTLS: fix $tls_out_ocsp under hosts_request_ocsp. Previously the
117 verification result was not updated unless hosts_require_ocsp applied.
119 JH/19 Bug 2398: fix listing of a named-queue. Previously, even with the option
120 queue_list_requires_admin set to false, non-admin users were denied the
123 JH/20 Bug 2389: fix server advertising of usable certificates, under GnuTLS in
124 directory-of-certs mode. Previously they were advertised despite the
127 JH/21 The smtp transport option "hosts_noproxy_tls" is now unset by default.
128 A single TCP connection by a client will now hold a TLS connection open
129 for multiple message deliveries, by default. Previoud the default was to
132 JH/22 The smtp transport option "hosts_try_dane" now enables all hosts by
133 default. If built with the facility, DANE will be used. The facility
134 SUPPORT_DANE is now enabled in the prototype build Makefile "EDITME".
136 JH/23 The build default is now for TLS to be included; the SUPPORT_TLS define
137 is replaced with DISABLE_TLS. Either USE_GNUTLS or (the new) USE_OPENSSL
138 must be defined and you must still, unless you define DISABLE_TLS, manage
139 the the include-dir and library-file requirements that go with that
140 choice. Non-TLS builds are still supported.
142 JH/24 Fix duplicated logging of peer name/address, on a transport connection-
145 JH/25 The smtp transport option "hosts_try_fastopen" now enables all hosts by
146 default. If the platform supports and has the facility enabled, it will
147 be requested on all coneections.
149 JH/26 The PIPE_CONNECT facility is promoted from experimental status and is now
150 controlled by the build-time option SUPPORT_PIPE_CONNECT.
152 PP/01 Unbreak heimdal_gssapi, broken in 4.92.
154 JH/27 Bug 2404: Use the main-section configuration option "dsn_from" for
155 success-DSN messages. Previously the From: header was always the default
156 one for these; the option was ignored.
158 JH/28 Fix the timeout on smtp response to apply to the whole response.
159 Previously it was reset for every read, so a teergrubing peer sending
160 single bytes within the time limit could extend the connection for a
161 long time. Credit to Qualsys Security Advisory Team for the discovery.
163 JH/29 Fix DSN Final-Recipient: field. Previously it was the post-routing
164 delivery address, which leaked information of the results of local
165 forwarding. Change to the original envelope recipient address, per
168 JH/30 Bug 2411: Fix DSN generation when RFC 3461 failure notification is
169 requested. Previously not bounce was generated and a log entry of
170 error ignored was made.
172 JH/31 Avoid re-expansion in ${sort } expansion. (CVE-2019-13917)
174 JH/32 Introduce a general tainting mechanism for values read from the input
175 channel, and values derived from them. Refuse to expand any tainted
176 values, to catch one form of exploit.
178 JH/33 Bug 2413: Fix dkim_strict option. Previously the expansion result
179 was unused and the unexpanded text used for the test. Found and
180 fixed by Ruben Jenster.
182 JH/34 Fix crash after TLS shutdown. When the TCP/SMTP channel was left open,
183 an attempt to use a TLS library read routine dereffed a nul pointer,
186 JH/35 Bug 2409: filter out-of-spec chars from callout response before using
187 them in our smtp response.
189 JH/36 Have the general router option retry_use_local_part default to true when
190 any of the restrictive preconditions are set (to anything). Previously it
191 was only for check_local user. The change removes one item of manual
192 configuration which is required for proper retries when a remote router
193 handles a subset of addresses for a domain.
195 JH/37 Appendfile: when evaluating quota use (non-quota_size_regex) take the file
196 link count into consideration.
198 HS/04 Fix handling of very log lines in -H files. If a -<key> <value> line
199 caused the extension of big_buffer, the following lines were ignored.
201 JH/38 Bug 1395: Teach the DNS negative-cache about TTL value from the SOA in
202 accordance with RFC 2308. Previously there was no expiry, so a longlived
203 receive process (eg. due to ACL delays) versus a short SOA value could
206 HS/05 Handle trailing backslash gracefully. (CVE-2019-15846)
208 JH/39 Promote DMARC support to mainline.
210 JH/40 Bug 2452: Add a References: header to DSNs.
212 JH/41 With GnuTLS 3.6.0 (and later) do not attempt to manage Diffie-Hellman
213 parameters. The relevant library call is documented as "Deprecated: This
214 function is unnecessary and discouraged on GnuTLS 3.6.0 or later. Since
215 3.6.0, DH parameters are negotiated following RFC7919."
217 JH/43 Bug 2465: Fix taint-handling in dsearch lookup. Previously a nontainted
218 buffer was used for the filename, resulting in a trap when tainted
219 arguments (eg. $domain) were used.
225 JH/01 Remove code calling the customisable local_scan function, unless a new
226 definition "HAVE_LOCAL_SCAN=yes" is present in the Local/Makefile.
228 JH/02 Bug 1007: Avoid doing logging from signal-handlers, as that can result in
229 non-signal-safe functions being used.
231 JH/03 Bug 2269: When presented with a received message having a stupidly large
232 number of DKIM-Signature headers, disable DKIM verification to avoid
233 a resource-consumption attack. The limit is set at twenty.
235 JH/04 Add variables $arc_domains, $arc_oldest_pass for ARC verify. Fix the
236 report of oldest_pass in ${authres } in consequence, and separate out
237 some descriptions of reasons for verification fail.
239 JH/05 Bug 2273: Cutthrough delivery left a window where the received messsage
240 files in the spool were present and unlocked. A queue-runner could spot
241 them, resulting in a duplicate delivery. Fix that by doing the unlock
242 after the unlink. Investigation by Tim Stewart. Take the opportunity to
243 add more error-checking on spoolfile handling while that code is being
246 PP/01 Refuse to open a spool data file (*-D) if it's a symlink.
247 No known attacks, no CVE, this is defensive hardening.
249 JH/06 Bug 2275: The MIME ACL unlocked the received message files early, and
250 a queue-runner could start a delivery while other operations were ongoing.
251 Cutthrough delivery was a common victim, resulting in duplicate delivery.
252 Found and investigated by Tim Stewart. Fix by using the open message data
253 file handle rather than opening another, and not locally closing it (which
254 releases a lock) for that case, while creating the temporary .eml format
255 file for the MIME ACL. Also applies to "regex" and "spam" ACL conditions.
257 JH/07 Bug 177: Make a random-recipient callout success visible in ACL, by setting
258 $sender_verify_failure/$recipient_verify_failure to "random".
260 JH/08 When generating a selfsigned cert, use serial number 1 since zero is not
263 JH/09 Bug 2274: Fix logging of cmdline args when starting in an unlinked cwd.
264 Previously this would segfault.
266 JH/10 Fix ARC signing for case when DKIM signing failed. Previously this would
269 JH/11 Bug 2264: Exim now only follows CNAME chains one step by default. We'd
270 like zero, since the resolver should be doing this for us, But we need one
271 as a CNAME but no MX presence gets the CNAME returned; we need to check
272 that doesn't point to an MX to declare it "no MX returned" rather than
273 "error, loop". A new main option is added so the older capability of
274 following some limited number of chain links is maintained.
276 JH/12 Add client-ip info to non-pass iprev ${authres } lines.
278 JH/13 For receent Openssl versions (1.1 onward) use modern generic protocol
279 methods. These should support TLS 1.3; they arrived with TLS 1.3 and the
280 now-deprecated earlier definitions used only specified the range up to TLS
281 1.2 (in the older-version library docs).
283 JH/14 Bug 2284: Fix DKIM signing for body lines starting with a pair of dots.
285 JH/15 Rework TLS client-side context management. Stop using a global, and
286 explicitly pass a context around. This enables future use of TLS for
287 connections to service-daemons (eg. malware scanning) while a client smtp
288 connection is using TLS; with cutthrough connections this is quite likely.
290 JH/16 Fix ARC verification to do AS checks in reverse order.
292 JH/17 Support a "tls" option on the ${readsocket } expansion item.
294 JH/18 Bug 2287: Fix the protocol name (eg utf8esmtp) for multiple messages
295 using the SMTPUTF8 option on their MAIL FROM commands, in one connection.
296 Previously the "utf8" would be re-prepended for every additional message.
298 JH/19 Reject MAIL FROM commands with SMTPUTF8 when the facility was not advertised.
299 Previously thery were accepted, resulting in issues when attempting to
300 forward messages to a non-supporting MTA.
302 PP/02 Let -n work with printing macros too, not just options.
304 JH/20 Bug 2296: Fix cutthrough for >1 address redirection. Previously only
305 one parent address was copied, and bogus data was used at delivery-logging
306 time. Either a crash (after delivery) or bogus log data could result.
307 Discovery and analysis by Tim Stewart.
309 PP/03 Make ${utf8clean:} expansion operator detect incomplete final character.
310 Previously if the string ended mid-character, we did not insert the
311 promised '?' replacement.
313 PP/04 Documentation: current string operators work on bytes, not codepoints.
315 JH/21 Change as many as possible of the global flags into one-bit bitfields; these
316 should pack well giving a smaller memory footprint so better caching and
317 therefore performance. Group the declarations where this can't be done so
318 that the byte-sized flag variables are not interspersed among pointer
319 variables, giving a better chance of good packing by the compiler.
321 JH/22 Bug 1896: Fix the envelope from for DMARC forensic reports to be possibly
322 non-null, to avoid issues with sites running BATV. Previously reports were
323 sent with an empty envelope sender so looked like bounces.
325 JH/23 Bug 2318: Fix the noerror command within filters. It wasn't working.
326 The ignore_error flag wasn't being returned from the filter subprocess so
327 was not set for later routers. Investigation and fix by Matthias Kurz.
329 JH/24 Bug 2310: Raise a msg:fail:internal event for each undelivered recipient,
330 and a msg:complete for the whole, when a message is manually removed using
331 -Mrm. Developement by Matthias Kurz, hacked on by JH.
333 JH/25 Avoid fixed-size buffers for pathnames in DB access. This required using
334 a "Gnu special" function, asprintf() in the DB utility binary builds; I
335 hope that is portable enough.
337 JH/26 Bug 2311: Fix DANE-TA verification under GnuTLS. Previously it was also
338 requiring a known-CA anchor certificate; make it now rely entirely on the
339 TLSA as an anchor. Checking the name on the leaf cert against the name
340 on the A-record for the host is still done for TA (but not for EE mode).
342 JH/27 Fix logging of proxy address. Previously, a pointless "PRX=[]:0" would be
343 included in delivery lines for non-proxied connections, when compiled with
344 SUPPORT_SOCKS and running with proxy logging enabled.
346 JH/28 Bug 2314: Fire msg:fail:delivery event even when error is being ignored.
347 Developement by Matthias Kurz, tweaked by JH. While in that bit of code,
348 move the existing event to fire before the normal logging of message
349 failure so that custom logging is bracketed by normal logging.
351 JH/29 Bug 2322: A "fail" command in a non-system filter (file) now fires the
352 msg:fail:internal event. Developement by Matthias Kurz.
354 JH/30 Bug 2329: Increase buffer size used for dns lookup from 2k, which was
355 far too small for todays use of crypto signatures stored there. Go all
356 the way to the max DNS message size of 64kB, even though this might be
357 overmuch for IOT constrained device use.
359 JH/31 Fix a bad use of a copy function, which could be used to pointlessly
360 copy a string over itself. The library routine is documented as not
361 supporting overlapping copies, and on MacOS it actually raised a SIGABRT.
363 JH/32 For main options check_spool_space and check_inode_space, where the
364 platform supports 64b integers, support more than the previous 2^31 kB
365 (i.e. more than 2 TB). Accept E, P and T multipliers in addition to
366 the previous G, M, k.
368 JH/33 Bug 2338: Fix the cyrus-sasl authenticator to fill in the
369 $authenticated_fail_id variable on authentication failure. Previously
372 JH/34 Increase RSA keysize of autogen selfsign cert from 1024 to 2048. RHEL 8.0
373 OpenSSL didn't want to use such a weak key. Do for GnuTLS also, and for
374 more-modern GnuTLS move from GNUTLS_SEC_PARAM_LOW to
375 GNUTLS_SEC_PARAM_MEDIUM.
377 JH/35 OpenSSL: fail the handshake when SNI processing hits a problem, server
378 side. Previously we would continue as if no SNI had been received.
380 JH/36 Harden the handling of string-lists. When a list consisted of a sole
381 "<" character, which should be a list-separator specification, we walked
382 off past the nul-terimation.
384 JH/37 Bug 2341: Send "message delayed" warning MDNs (restricted to external
385 causes) even when the retry time is not yet met. Previously they were
386 not, meaning that when (say) an account was over-quota and temp-rejecting,
387 and multiple senders' messages were queued, only one sender would get
388 notified on each configured delay_warning cycle.
390 JH/38 Bug 2351: Log failures to extract envelope addresses from message headers.
392 JH/39 OpenSSL: clear the error stack after an SSL_accept(). With anon-auth
393 cipher-suites, an error can be left on the stack even for a succeeding
394 accept; this results in impossible error messages when a later operation
397 AM/01 Bug 2359: GnuTLS: repeat lowlevel read and write operations while they
398 return error codes indicating retry. Under TLS1.3 this becomes required.
400 JH/40 Fix the feature-cache refresh for EXPERIMENTAL_PIPE_CONNECT. Previously
401 it only wrote the new authenticators, resulting in a lack of tracking of
402 peer changes of ESMTP extensions until the next cache flush.
404 JH/41 Fix the loop reading a message header line to check for integer overflow,
405 and more-often against header_maxsize. Previously a crafted message could
406 induce a crash of the recive process; now the message is cleanly rejected.
408 JH/42 Bug 2366: Fix the behaviour of the dkim_verify_signers option. It had
409 been totally disabled for all of 4.91. Discovery and fix by "Mad Alex".
415 GF/01 DEFER rather than ERROR on redis cluster MOVED response.
416 When redis_servers is set to a list of > 1 element, and the Redis servers
417 in that list are in cluster configuration, convert the REDIS_REPLY_ERROR
418 case of MOVED into a DEFER case instead, thus moving the query onto the
419 next server in the list. For a cluster of N elements, all N servers must
420 be defined in redis_servers.
422 GF/02 Catch and remove uninitialized value warning in exiqsumm
423 Check for existence of @ARGV before looking at $ARGV[0]
425 JH/01 Replace the store_release() internal interface with store_newblock(),
426 which internalises the check required to safely use the old one, plus
427 the allocate and data copy operations duplicated in both (!) of the
428 extant use locations.
430 JH/02 Disallow '/' characters in queue names specified for the "queue=" ACL
431 modifier. This matches the restriction on the commandline.
433 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
434 Previously only the last row was returned.
436 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
437 we assumed that tags in the header were well-formed, and parsed the
438 element content after inspecting only the first char of the tag.
439 Assumptions at that stage could crash the receive process on malformed
442 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
443 While running the DKIM ACL we operate on the Permanent memory pool so that
444 variables created with "set" persist to the DATA ACL. Also (at any time)
445 DNS lookups that fail create cache records using the Permanent pool. But
446 expansions release any allocations made on the current pool - so a dnsdb
447 lookup expansion done in the DKIM ACL releases the memory used for the
448 DNS negative-cache, and bad things result. Solution is to switch to the
449 Main pool for expansions.
450 While we're in that code, add checks on the DNS cache during store_reset,
451 active in the testsuite.
452 Problem spotted, and debugging aided, by Wolfgang Breyha.
454 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
455 When none of the hosts presented to a transport match an already-open
456 connection, close it and proceed with the list. Previously we would
457 queue the message. Spotted by Lena with Yahoo, probably involving
460 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
461 Previously a spurious "250 OK id=" response was appended to the proper
464 JH/08 The "support for" informational output now, which built with Content
465 Scanning support, has a line for the malware scanner interfaces compiled
466 in. Interface can be individually included or not at build time.
468 JH/09 The "aveserver", "kavdaemon" and "mksd" interfaces are now not included
469 by the template makefile "src/EDITME". The "STREAM" support for an older
470 ClamAV interface method is removed.
472 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
473 rows affected is given instead).
475 JH/11 The runtime Berkeley DB library version is now additionally output by
476 "exim -d -bV". Previously only the compile-time version was shown.
478 JH/12 Bug 2230: Fix cutthrough routing for nonfirst messages in an initiating
479 SMTP connection. Previously, when one had more receipients than the
480 first, an abortive onward connection was made. Move to full support for
481 multiple onward connections in sequence, handling cutthrough connection
482 for all multi-message initiating connections.
484 JH/13 Bug 2229: Fix cutthrough routing for nonstandard port numbers defined by
485 routers. Previously, a multi-recipient message would fail to match the
486 onward-connection opened for the first recipient, and cause its closure.
488 JH/14 Bug 2174: A timeout on connect for a callout was also erroneously seen as
489 a timeout on read on a GnuTLS initiating connection, resulting in the
490 initiating connection being dropped. This mattered most when the callout
491 was marked defer_ok. Fix to keep the two timeout-detection methods
494 JH/15 Relax results from ACL control request to enable cutthrough, in
495 unsupported situations, from error to silently (except under debug)
496 ignoring. This covers use with PRDR, frozen messages, queue-only and
499 HS/01 Fix Buffer overflow in base64d() (CVE-2018-6789)
501 JH/16 Fix bug in DKIM verify: a buffer overflow could corrupt the malloc
502 metadata, resulting in a crash in free().
504 PP/01 Fix broken Heimdal GSSAPI authenticator integration.
505 Broken in f2ed27cf5, missing an equals sign for specified-initialisers.
506 Broken also in d185889f4, with init system revamp.
508 JH/17 Bug 2113: Fix conversation closedown with the Avast malware scanner.
509 Previously we abruptly closed the connection after reading a malware-
510 found indication; now we go on to read the "scan ok" response line,
513 JH/18 Bug 2239: Enforce non-usability of control=utf8_downconvert in the mail
514 ACL. Previously, a crash would result.
516 JH/19 Speed up macro lookups during configuration file read, by skipping non-
517 macro text after a replacement (previously it was only once per line) and
518 by skipping builtin macros when searching for an uppercase lead character.
520 JH/20 DANE support moved from Experimental to mainline. The Makefile control
521 for the build is renamed.
523 JH/21 Fix memory leak during multi-message connections using STARTTLS. A buffer
524 was allocated for every new TLS startup, meaning one per message. Fix
525 by only allocating once (OpenSSL) or freeing on TLS-close (GnuTLS).
527 JH/22 Bug 2236: When a DKIM verification result is overridden by ACL, DMARC
528 reported the original. Fix to report (as far as possible) the ACL
529 result replacing the original.
531 JH/23 Fix memory leak during multi-message connections using STARTTLS under
532 OpenSSL. Certificate information is loaded for every new TLS startup,
533 and the resources needed to be freed.
535 JH/24 Bug 2242: Fix exim_dbmbuild to permit directoryless filenames.
537 JH/25 Fix utf8_downconvert propagation through a redirect router. Previously it
540 JH/26 Bug 2253: For logging delivery lines under PRDR, append the overall
541 DATA response info to the (existing) per-recipient response info for
542 the "C=" log element. It can have useful tracking info from the
543 destination system. Patch from Simon Arlott.
545 JH/27 Bug 2251: Fix ldap lookups that return a single attribute having zero-
546 length value. Previously this would segfault.
548 HS/02 Support Avast multiline protoocol, this allows passing flags to
549 newer versions of the scanner.
551 JH/28 Ensure that variables possibly set during message acceptance are marked
552 dead before release of memory in the daemon loop. This stops complaints
553 about them when the debug_store option is enabled. Discovered specifically
554 for sender_rate_period, but applies to a whole set of variables.
555 Do the same for the queue-runner and queue-list loops, for variables set
556 from spool message files. Do the same for the SMTP per-message loop, for
557 certain variables indirectly set in ACL operations.
559 JH/29 Bug 2250: Fix a longstanding bug in heavily-pipelined SMTP input (such
560 as a multi-recipient message from a mailinglist manager). The coding had
561 an arbitrary cutoff number of characters while checking for more input;
562 enforced by writing a NUL into the buffer. This corrupted long / fast
563 input. The problem was exposed more widely when more pipelineing of SMTP
564 responses was introduced, and one Exim system was feeding another.
565 The symptom is log complaints of SMTP syntax error (NUL chars) on the
566 receiving system, and refused recipients seen by the sending system
567 (propating to people being dropped from mailing lists).
568 Discovered and pinpointed by David Carter.
570 JH/30 The (EXPERIMENTAL_DMARC) variable $dmarc_ar_header is withdrawn, being
571 replaced by the ${authresults } expansion.
573 JH/31 Bug 2257: Fix pipe transport to not use a socket-only syscall.
575 HS/03 Set a handler for SIGTERM and call exit(3) if running as PID 1. This
576 allows proper process termination in container environments.
578 JH/32 Bug 2258: Fix spool_wireformat in combination with LMTP transport.
579 Previously the "final dot" had a newline after it; ensure it is CR,LF.
581 JH/33 SPF: remove support for the "spf" ACL condition outcome values "err_temp"
582 and "err_perm", deprecated since 4.83 when the RFC-defined words
583 "temperror" and "permerror" were introduced.
585 JH/34 Re-introduce enforcement of no cutthrough delivery on transports having
586 transport-filters or DKIM-signing. The restriction was lost in the
587 consolidation of verify-callout and delivery SMTP handling.
588 Extend the restriction to also cover ARC-signing.
590 JH/35 Cutthrough: for a final-dot response timeout (and nonunderstood responses)
591 in defer=pass mode supply a 450 to the initiator. Previously the message
594 PP/02 DANE: add dane_require_tls_ciphers SMTP Transport option; if unset,
595 tls_require_ciphers is used as before.
597 HS/03 Malware Avast: Better match the Avast multiline protocol. Add
598 "pass_unscanned". Only tmpfails from the scanner are written to
599 the paniclog, as they may require admin intervention (permission
600 denied, license issues). Other scanner errors (like decompression
601 bombs) do not cause a paniclog entry.
603 JH/36 Fix reinitialisation of DKIM logging variable between messages.
604 Previously it was possible to log spurious information in receive log
607 JH/37 Bug 2255: Revert the disable of the OpenSSL session caching. This
608 triggered odd behaviour from Outlook Express clients.
610 PP/03 Add util/renew-opendmarc-tlds.sh script for safe renewal of public
613 JH/38 DKIM: accept Ed25519 pubkeys in SubjectPublicKeyInfo-wrapped form,
614 since the IETF WG has not yet settled on that versus the original
615 "bare" representation.
617 JH/39 Fix syslog logging for syslog_timestamp=no and log_selector +millisec.
618 Previously the millisecond value corrupted the output.
619 Fix also for syslog_pid=no and log_selector +pid, for which the pid
620 corrupted the output.
626 JH/01 Rework error string handling in TLS interface so that the caller in
627 more cases is responsible for logging. This permits library-sourced
628 string to be attached to addresses during delivery, and collapses
629 pairs of long lines into single ones.
631 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
632 during configuration. Wildcards are allowed and expanded.
634 JH/02 Rework error string handling in DKIM to pass more info back to callers.
635 This permits better logging.
637 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
638 do not close it down and have the child transport start it up again on
639 the passed-on TCP connection. Instead, proxy the child (and any
640 subsequent ones) for TLS via a unix-domain socket channel. Logging is
641 affected: the continued delivery log lines do not have any DNSSEC, TLS
642 Certificate or OCSP information. TLS cipher information is still logged.
644 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
645 identical IP addresses on different listening ports. Will also affect
648 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
649 add noisy ifdef guards to special-case this sillyness.
650 Patch from Bernd Kuhls.
652 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
653 than 255 are no longer allowed.
655 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
656 Disable the session-cache too, which might reduce our load. Since we
657 currrectly use a new context for every connection, both as server and
658 client, there is no benefit for these.
659 GnuTLS appears to not support tickets server-side by default (we don't
660 call gnutls_session_ticket_enable_server()) but client side is enabled
661 by default on recent versions (3.1.3 +) unless the PFS priority string
664 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
665 <https://reproducible-builds.org/specs/source-date-epoch/>.
667 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
668 the check for any unsuccessful recipients did not notice the limit, and
669 erroneously found still-pending ones.
671 JH/08 Pipeline CHUNKING command and data together, on kernels that support
672 MSG_MORE. Only in-clear (not on TLS connections).
674 JH/09 Avoid using a temporary file during transport using dkim. Unless a
675 transport-filter is involved we can buffer the headers in memory for
676 creating the signature, and read the spool data file once for the
677 signature and again for transmission.
679 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
680 4.77 as the kernel support then wasn't solid, having issues in 64bit
681 mode. Now, it's been long enough. Add support for FreeBSD also.
683 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
684 case where the routing stage had gathered several addresses to send to
685 a host before calling the transport for the first, we previously failed
686 to close down TLS in the old transport process before passing the TCP
687 connection to the new process. The new one sent a STARTTLS command
688 which naturally failed, giving a failed delivery and bloating the retry
689 database. Investigation and fix prototype from Wolfgang Breyha.
691 JH/12 Fix check on SMTP command input synchronisation. Previously there were
692 false-negatives in the check that the sender had not preempted a response
693 or prompt from Exim (running as a server), due to that code's lack of
694 awareness of the SMTP input buffering.
696 PP/04 Add commandline_checks_require_admin option.
697 Exim drops privileges sanely, various checks such as -be aren't a
698 security problem, as long as you trust local users with access to their
699 own account. When invoked by services which pass untrusted data to
700 Exim, this might be an issue. Set this option in main configuration
701 AND make fixes to the calling application, such as using `--` to stop
704 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
705 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
706 responses to those, into a single TLS record each way (this usually means
707 a single packet). As a side issue, smtp_enforce_sync now works on TLS
710 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
711 affects you only if you're dancing at the edge of the param size limits.
712 If you are, and this message makes sense to you, then: raise the
713 configured limit or use OpenSSL 1.1. Nothing we can do for older
716 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
717 cmdline element to get the documented default one. Previously it was
720 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
721 are unsuable for non-message-specific reasons - eg. connection timeout,
722 banner-time rejection.
724 JH/16 Fix logging of delivery remote port, when specified by router, under
727 PP/06 Repair manualroute's ability to take options in any order, even if one
728 is the name of a transport.
731 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
733 JH/17 Change the list-building routines interface to use the expanding-string
734 triplet model, for better allocation and copying behaviour.
736 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
737 Previously it was constructed the first time a possibly-matching string
738 was met in the configuration file input during startup; now it is done
741 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
742 compatible one, to avoid the (poorly documented) possibility of a config
743 file in the working directory redirecting the DB files, possibly correpting
744 some existing file. CVE-2017-10140 assigned for BDB.
746 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
747 cache-hot. Previously, although the result was properly cached, the
748 initial verify call returned a defer.
750 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
751 the main verify for receipient in uncached-mode.
753 JH/22 Retire historical build files to an "unsupported" subdir. These are
754 defined as "ones for which we have no current evidence of testing".
756 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
757 if present. Previously it was ignored.
759 JH/24 Start using specified-initialisers in C structure init coding. This is
760 a C99 feature (it's 2017, so now considered safe).
762 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
763 if was a fixed-sized field and bitmask ops via macros; it is now more
766 PP/07 GitHub PR 56: Apply MariaDB build fix.
767 Patch provided by Jaroslav Škarvada.
769 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
770 during Coverity cleanups [4.87 JH/47]
771 Diagnosis and fix provided by Michael Fischer v. Mollard.
773 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
774 the right size to place the terminating semicolon on its own folded
775 line, the header hash was calculated to an incorrect value thanks to
776 the (relaxed) space the fold became.
778 HS/02 Fix Bug 2130: large writes from the transport subprocess were chunked
779 and confused the parent.
781 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
782 which could crash as a result. This could lead to undeliverable messages.
784 JH/28 Logging: "next input sent too soon" now shows where input was truncated
787 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
788 matters on fast-turnover and PID-randomising systems, which were getting
789 out-of-order delivery.
791 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
792 a possibly-overlapping copy. The symptom was that "Remote host closed
793 connection in response to HELO" was logged instead of the actual 4xx
796 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
797 Previously only that bufferd was discarded, resulting in SYMTP command
800 JH/32 DKIM: when a message has multiple signatures matching an identity given
801 in dkim_verify_signers, run the dkim acl once for each. Previously only
802 one run was done. Bug 2189.
804 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
805 "panic the current process" to "deliberately defer". The panic log is
806 still written with the problem list name; the mail and reject logs now
807 get a temp-reject line for the message that was being handled, saying
808 something like "domains check lookup or other defer". The SMTP 451
809 message is still "Temporary local problem".
811 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
812 A crafted sequence of BDAT commands could result in in-use memory beeing
813 freed. CVE-2017-16943.
815 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
816 from SMTP input. Previously it was always done; now only done for DATA
817 and not BDAT commands. CVE-2017-16944.
819 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
820 to the message (such as an overlong header line). Previously this was
821 not done and we did not exit BDAT mode. Followon from the previous item
822 though a different problem.
828 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
829 than -2003 did; needs libidn2 in addition to libidn.
831 JH/02 The path option on a pipe transport is now expanded before use.
833 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
834 Patch provided by "Björn", documentation fix added too.
836 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
837 missing a wire-to-host endian conversion.
839 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
840 close after a BDAT command line could be taken as a following command,
841 giving a synch failure. Fix by only checking for synch immediately
842 before acknowledging the chunk.
844 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
845 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
846 macro. Patches provided by Josh Soref.
848 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
849 Previously we did not; the RFC seems ambiguous and VRFY is not listed
850 by IANA as a service extension. However, John Klensin suggests that we
853 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
854 the dkim code may be unix-mode line endings rather than smtp wire-format
855 CRLF, so prepend a CR to any bare LF.
857 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
858 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
860 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
861 the first were themselves being wrongly included in the feed into dkim
862 processing; with most chunk sizes in use this resulted in an incorrect
863 body hash calculated value.
865 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
866 DKIM signature block, for verification. Although advised against by
867 standards it is specifically not ruled illegal.
869 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
871 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
872 missing a body hash (the bh= tag).
874 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
875 It seems that HAProxy sends the Proxy Protocol information in clear and
876 only then does a TLS startup, so do the same.
878 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
879 TCP connections (such as for Spamd) unless the daemon successfully set
880 Fast Open mode on its listening sockets. This fixes breakage seen on
881 too-old kernels or those not configured for Fast Open, at the cost of
882 requiring both directions being enabled for TFO, and TFO never being used
883 by non-daemon-related Exim processes.
885 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
886 endings, at least on the first header line. Try to canonify any that get
887 past that check, despite the cost.
889 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
890 now limited to an arbitrary five deep, while parsing addresses with the
891 strip_excess_angle_brackets option enabled.
893 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
894 instead leave the unprompted TLS handshake in socket buffer for the
895 TLS library to consume.
897 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
899 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
901 JH/16 Drop variables when they go out of scope. Memory management drops a whole
902 region in one operation, for speed, and this leaves assigned pointers
903 dangling. Add checks run only under the testsuite which checks all
904 variables at a store-reset and panics on a dangling pointer; add code
905 explicitly nulling out all the variables discovered. Fixes one known
906 bug: a transport crash, where a dangling pointer for $sending_ip_address
907 originally assigned in a verify callout, is re-used.
909 PP/06 Drop '.' from @INC in various Perl scripts.
911 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
913 PP/08 Reduce a number of compilation warnings under clang; building with
914 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
915 should be warning-free.
917 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
919 HS/01 Fix portability problems introduced by PP/08 for platforms where
920 realloc(NULL) is not equivalent to malloc() [SunOS et al].
922 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
923 chunk. This allows us to accept broken chunked messages. We need a more
924 general solution here.
926 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
927 already-broken messages in the queue.
929 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
931 JH/19 Fix reference counting bug in routing-generated-address tracking.
937 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
938 supports it and a size is available (ie. the sending peer gave us one).
940 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
941 years of being deprecated). The replacements are the ACLs
942 acl_smtp_mime and acl_not_smtp_mime.
944 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
945 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
946 or even an in-clear connection were permitted. Now, if the host lookup
947 was dnssec and dane was requested then the host is only used if the
948 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
949 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
950 if one fails this test.
951 This means that a poorly-configured remote DNS will make it incommunicado;
952 but it protects against a DNS-interception attack on it.
954 JH/04 Bug 1810: make continued-use of an open smtp transport connection
955 non-noisy when a race steals the message being considered.
957 JH/05 If main configuration option tls_certificate is unset, generate a
958 self-signed certificate for inbound TLS connections.
960 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
961 in rewrites and routers.
963 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
964 and logged a warning sing 4.83; now they are a configuration file error.
966 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
967 (lacking @domain). Apply the same qualification processing as RCPT.
969 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
971 JH/10 Support ${sha256:} applied to a string (as well as the previous
974 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
975 a cutthrough deliver is pending, as we always want to make a connection.
976 This also avoids re-routing the message when later placing the cutthrough
977 connection after a verify cache hit.
978 Do not update it with the verify result either.
980 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
981 when routing results in more than one destination address.
983 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
984 signing (which inhibits the cutthrough capability). Previously only
985 the presence of an option was tested; now an expansion evaluating as
986 empty is permissible (obviously it should depend only on data available
987 when the cutthrough connection is made).
989 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
990 the relevant preceding SMTP command did not note the pipelining mode.
992 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
993 Previously they were not counted.
995 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
996 as one having no matching records. Previously we deferred the message
997 that needed the lookup.
999 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
1000 distinguished as "(=".
1002 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
1003 for missing MX records. Previously it only worked for missing A records.
1005 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
1007 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
1008 after the data-go-ahead and data-ack. Patch from Jason Betts.
1010 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
1011 even for a "none" policy. Patch from Tony Meyer.
1013 JH/22 Fix continued use of a connection for further deliveries. If a port was
1014 specified by a router, it must also match for the delivery to be
1017 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
1018 When one of the recipients of a message was unsuitable for the connection
1019 (has no matching addresses), we lost track of needing to mark it
1020 deferred. As a result mail would be lost.
1022 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
1024 JH/25 Decoding ACL controls is now done using a binary search; the source code
1025 takes up less space and should be simpler to maintain. Merge the ACL
1026 condition decode tables also, with similar effect.
1028 JH/26 Fix problem with one_time used on a redirect router which returned the
1029 parent address unchanged. A retry would see the parent address marked as
1030 delivered, so not attempt the (identical) child. As a result mail would
1033 JH/27 Fix a possible security hole, wherein a process operating with the Exim
1034 UID can gain a root shell. Credit to http://www.halfdog.net/ for
1035 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
1038 JH/28 Enable {spool,log} filesystem space and inode checks as default.
1039 Main config options check_{log,spool}_{inodes,space} are now
1040 100 inodes, 10MB unless set otherwise in the configuration.
1042 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
1043 Previously it only applied to the main-section connection policy
1046 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
1048 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
1049 by me. Added RFC7919 DH primes as an alternative.
1051 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
1052 are not in the system include path.
1054 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
1055 GnuTLS, when a session startup failed (eg because the client disconnected)
1056 Exim did stdio operations after fclose. This was exposed by a recent
1057 change which nulled out the file handle after the fclose.
1059 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
1060 signed directly by the cert-signing cert, rather than an intermediate
1061 OCSP-signing cert. This is the model used by LetsEncrypt.
1063 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
1065 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
1066 an incoming connection.
1068 HS/02 Bug 1802: Do not half-close the connection after sending a request
1071 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
1072 fallback to "prime256v1".
1074 JH/34 SECURITY: Use proper copy of DATA command in error message.
1075 Could leak key material. Remotely exploitable. CVE-2016-9963.
1081 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
1082 and 3.4.4 - once the server is enabled to respond to an OCSP request
1083 it does even when not requested, resulting in a stapling non-aware
1084 client dropping the TLS connection.
1086 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
1087 support variable-length bit vectors. No functional change.
1089 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
1090 The I= interface field on outgoing lines is now after the H= remote
1091 host field, same as incoming lines. There is a separate
1092 outgoing_interface log selector which allows you to disable the
1095 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
1096 If not running log_selector +smtp_connection the mainlog would be held
1097 open indefinitely after a "too many connections" event, including to a
1098 deleted file after a log rotate. Leave the per net connection logging
1099 leaving it open for efficiency as that will be quickly detected by the
1100 check on the next write.
1102 HS/01 Bug 1671: Fix post transport crash.
1103 Processing the wait-<transport> messages could crash the delivery
1104 process if the message IDs didn't exist for some reason. When
1105 using 'split_spool_directory=yes' the construction of the spool
1106 file name failed already, exposing the same netto behaviour.
1108 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
1109 mime_regex ACL conditions.
1111 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
1112 to DSN fail messages (bounces): remote IP, remote greeting, remote response
1113 to HELO, local diagnostic string.
1115 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
1116 log line to debug. Even when configured with a tls authenticator many
1117 client connections are expected to not authenticate in this way, so
1118 an authenticate fail is not an error.
1120 HS/02 Add the Exim version string to the process info. This way exiwhat
1121 gives some more detail about the running daemon.
1123 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
1124 matter for fast-change records such as DNSBLs.
1126 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
1127 retry record, even if constant. There may be multiple transports with
1128 different interface settings and the retry behaviour needs to be kept
1131 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
1133 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
1135 JH/10 Bug 840: fix log_defer_output option of pipe transport
1137 JH/11 Bug 830: use same host for all RCPTS of a message, even under
1138 hosts_randomize. This matters a lot when combined with mua_wrapper.
1140 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
1141 ${quote_pgsql:<string>} operator.
1143 JH/13 Bug 1708: avoid misaligned access in cached lookup.
1145 JH/14 Change header file name for freeradius-client. Relevant if compiling
1146 with Radius support; from the Gentoo tree and checked under Fedora.
1148 JH/15 Bug 1712: Introduce $prdr_requested flag variable
1150 JH/16 Bug 1714: Permit an empty string as expansion result for transport
1151 option transport_filter, meaning no filtering.
1153 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
1155 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
1156 defaults to "*" (all hosts). The variable is now available when not built
1157 with TLS, default unset, mainly to enable keeping the testsuite sane.
1158 If a server certificate is not supplied (via tls_certificate) an error is
1159 logged, and clients will find TLS connections fail on startup. Presumably
1160 they will retry in-clear.
1161 Packagers of Exim are strongly encouraged to create a server certificate
1162 at installation time.
1164 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
1165 with the $config_file variable.
1167 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
1168 in transport context, after the attempt, and per-recipient. The latter type
1169 is per host attempted. The event data is the error message, and the errno
1170 information encodes the lookup type (A vs. MX) used for the (first) host,
1171 and the trailing two digits of the smtp 4xx response.
1173 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
1174 to write to mainlog (or rejectlog, paniclog) in the window between file
1175 creation and permissions/ownership being changed. Particularly affects
1176 installations where exicyclog is run as root, rather than exim user;
1177 result is that the running daemon panics and dies.
1179 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
1181 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
1182 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
1183 "pri" and "weight". Note that the previous implicit priority given by the
1184 list order is no longer honoured.
1186 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
1187 for DKIM processing.
1189 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
1190 by defining SUPPORT_SOCKS.
1192 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
1193 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
1194 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
1195 variables are renamed to proxy_{local,external}_{address,port}.
1197 JH/27 Move Internationalisation support from Experimental to mainline, enabled
1198 for a build by defining SUPPORT_I18N
1200 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
1201 of the query string, and make ${quote_redis:} do that quoting.
1203 JH/29 Move Events support from Experimental to mainline, enabled by default
1204 and removable for a build by defining DISABLE_EVENT.
1206 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
1208 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
1209 cached by the daemon.
1211 JH/32 Move Redis support from Experimental to mainline, enabled for a build
1212 by defining LOOKUP_REDIS. The libhiredis library is required.
1214 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
1215 keys are given for lookup.
1217 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
1218 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
1219 only supported when built with TLS support. The PolarSSL SHA routines
1220 are still used when the TLS library is too old for convenient support.
1222 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
1223 openssl_options), for security. OpenSSL forces this from version 1.1.0
1224 server-side so match that on older versions.
1226 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
1227 allocation for $value could be released as the expansion processing
1228 concluded, but leaving the global pointer active for it.
1230 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
1231 and to use the domains and local_parts ACL conditions.
1233 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
1234 incorrectly not doubled on cutthrough transmission, hence seen as a
1235 body-termination at the receiving system - resulting in truncated mails.
1236 Commonly the sender saw a TCP-level error, and retransmitted the message
1237 via the normal store-and-forward channel. This could result in duplicates
1238 received - but deduplicating mailstores were liable to retain only the
1239 initial truncated version.
1241 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
1243 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
1245 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
1246 we're in there, support oversigning also; bug 1309.
1248 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
1250 HS/04 Add support for keep_environment and add_environment options.
1252 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
1253 either intentional arithmetic overflow during PRNG, or testing config-
1256 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
1257 delivery resulted in actual delivery. Cancel cutthrough before DATA
1260 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
1261 rejects a recipient: pass the reject to the originator.
1263 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
1264 Many were false-positives and ignorable, but it's worth fixing the
1267 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
1268 for the new environment-manipulation done at startup. Move the routines
1269 from being local to tls.c to being global via the os.c file.
1271 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
1272 an extract embedded as result-arg for a map, the first arg for extract
1273 is unavailable so we cannot tell if this is a numbered or keyed
1274 extraction. Accept either.
1280 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
1283 JH/02 The smtp transport option "multi_domain" is now expanded.
1285 JH/03 The smtp transport now requests PRDR by default, if the server offers
1288 JH/04 Certificate name checking on server certificates, when exim is a client,
1289 is now done by default. The transport option tls_verify_cert_hostnames
1290 can be used to disable this per-host. The build option
1291 EXPERIMENTAL_CERTNAMES is withdrawn.
1293 JH/05 The value of the tls_verify_certificates smtp transport and main options
1294 default to the word "system" to access the system default CA bundle.
1295 For GnuTLS, only version 3.0.20 or later.
1297 JH/06 Verification of the server certificate for a TLS connection is now tried
1298 (but not required) by default. The verification status is now logged by
1299 default, for both outbound TLS and client-certificate supplying inbound
1302 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
1305 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
1306 Status Notification (bounce) messages are now MIME format per RFC 3464.
1307 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
1308 under the control of the dsn_advertise_hosts option, and routers may
1309 have a dsn_lasthop option.
1311 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
1312 default, modifiable by a malware= option. The list separator for
1313 the options can now be changed in the usual way. Bug 68.
1315 JH/10 The smtp_receive_timeout main option is now expanded before use.
1317 JH/11 The incoming_interface log option now also enables logging of the
1318 local interface on delivery outgoing connections.
1320 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
1321 if the interface and destination host and port all match.
1323 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
1326 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
1327 Patch from Andrew Lewis.
1329 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
1330 now supports optional time-restrictions, weighting, and priority
1331 modifiers per server. Patch originally by <rommer@active.by>.
1333 JH/16 The spamd_address main option now supports a mixed list of local
1334 and remote servers. Remote servers can be IPv6 addresses, and
1335 specify a port-range.
1337 JH/17 Bug 68: The spamd_address main option now supports an optional
1338 timeout value per server.
1340 JH/18 Bug 1581: Router and transport options headers_add/remove can
1341 now have the list separator specified.
1343 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
1346 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
1349 JH/21 Support for the A6 type of dns record is withdrawn.
1351 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
1352 rather than the verbs used.
1354 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
1355 from 255 to 1024 chars.
1357 JH/24 Verification callouts now attempt to use TLS by default.
1359 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
1360 are generic router options now. The defaults didn't change.
1362 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
1363 Original patch from Alexander Shikoff, worked over by JH.
1365 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
1366 files if ZCAT_COMMAND is not executable.
1368 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
1370 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
1372 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
1373 Normally benign, it bites when the pair was led to by a CNAME;
1374 modern usage is to not canonicalize the domain to a CNAME target
1375 (and we were inconsistent anyway for A-only vs AAAA+A).
1377 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
1379 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
1380 when evaluating $sender_host_dnssec.
1382 JH/31 Check the HELO verification lookup for DNSSEC, adding new
1383 $sender_helo_dnssec variable.
1385 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
1387 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
1389 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
1391 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
1392 documented as working, but never had. Support all but $spam_report.
1394 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
1395 added for tls authenticator.
1397 HS/03 Add perl_taintmode main config option
1403 TL/01 When running the test suite, the README says that variables such as
1404 no_msglog_check are global and can be placed anywhere in a specific
1405 test's script, however it was observed that placement needed to be near
1406 the beginning for it to behave that way. Changed the runtest perl
1407 script to read through the entire script once to detect and set these
1408 variables, reset to the beginning of the script, and then run through
1409 the script parsing/test process like normal.
1411 TL/02 The BSD's have an arc4random API. One of the functions to induce
1412 adding randomness was arc4random_stir(), but it has been removed in
1413 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
1414 function when detected.
1416 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
1417 cause callback expansion.
1419 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
1420 syntax errors in an expansion can be treated as a string instead of
1421 logging or causing an error, due to the internal use of bool_lax
1422 instead of bool when processing it.
1424 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
1425 server certificates when making smtp deliveries.
1427 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
1429 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
1431 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
1433 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
1434 Merged patch from Sebastian Wiedenroth.
1436 JH/05 Fix results-pipe from transport process. Several recipients, combined
1437 with certificate use, exposed issues where response data items split
1438 over buffer boundaries were not parsed properly. This eventually
1439 resulted in duplicates being sent. This issue only became common enough
1440 to notice due to the introduction of connection certificate information,
1441 the item size being so much larger. Found and fixed by Wolfgang Breyha.
1443 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
1444 size buffer was used, resulting in syntax errors when an expansion
1447 JH/07 Add support for directories of certificates when compiled with a GnuTLS
1448 version 3.3.6 or later.
1450 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
1451 is EXPERIMENTAL_EVENT, the main-configuration and transport options
1452 both become "event_action", the variables become $event_name, $event_data
1453 and $event_defer_errno. There is a new variable $verify_mode, usable in
1454 routers, transports and related events. The tls:cert event is now also
1455 raised for inbound connections, if the main configuration event_action
1458 TL/06 In test suite, disable OCSP for old versions of openssl which contained
1459 early OCSP support, but no stapling (appears to be less than 1.0.0).
1461 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
1462 server certificate names available under the smtp transport option
1463 "tls_verify_cert_hostname" now do not permit multi-component wildcard
1466 JH/10 Time-related extraction expansions from certificates now use the main
1467 option "timezone" setting for output formatting, and are consistent
1468 between OpenSSL and GnuTLS compilations. Bug 1541.
1470 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
1471 encoded parameter in the incoming message. Bug 1558.
1473 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
1474 include certificate info, eximon was claiming there were spoolfile
1477 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
1479 JH/14 Log delivery-related information more consistently, using the sequence
1480 "H=<name> [<ip>]" wherever possible.
1482 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
1483 are problematic for Debian distribution, omit them from the release
1486 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
1488 JH/16 Fix string representation of time values on 64bit time_t architectures.
1491 JH/17 Fix a null-indirection in certextract expansions when a nondefault
1492 output list separator was used.
1497 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
1498 checkers that were complaining about end of non-void function with no
1501 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
1502 This was a regression introduced in 4.83 by another bugfix.
1504 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1506 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1507 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1513 TF/01 Correctly close the server side of TLS when forking for delivery.
1515 When a message was received over SMTP with TLS, Exim failed to clear up
1516 the incoming connection properly after forking off the child process to
1517 deliver the message. In some situations the subsequent outgoing
1518 delivery connection happened to have the same fd number as the incoming
1519 connection previously had. Exim would try to use TLS and fail, logging
1520 a "Bad file descriptor" error.
1522 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1523 utilities have not been installed.
1525 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1526 temporary space as the ACL may create new global variables.
1528 TL/01 LDAP support uses per connection or global context settings, depending
1529 upon the detected version of the libraries at build time.
1531 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1532 to extract and use the src ip:port in logging and expansions as if it
1533 were a direct connection from the outside internet. PPv2 support was
1534 updated based on HAProxy spec change in May 2014.
1536 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1538 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1539 Properly escape header and check for NULL return.
1541 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1544 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1546 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1547 characters in header names, implemented as a verify condition.
1548 Contributed by Michael Fischer v. Mollard.
1550 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1551 results permerror and temperror. Previous values are deprecated but
1552 still accepted. In a future release, err_perm and err_temp will be
1553 completely removed, which will be a backward incompatibility if the
1554 ACL tests for either of these two old results. Patch contributed by
1555 user bes-internal on the mailing list.
1557 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1559 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1560 selectors, in both main and reject logs.
1562 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1565 JH/07 Add malware type "sock" for talking to simple daemon.
1567 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1569 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1570 routers/transports under cutthrough routing.
1572 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1573 numbers. Touch up "bool" conditional to keep the same definition.
1575 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1577 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1579 JH/12 Expand items in router/transport headers_add or headers_remove lists
1580 individually rather than the list as a whole. Bug 1452.
1582 Required for reasonable handling of multiple headers_ options when
1583 they may be empty; requires that headers_remove items with embedded
1584 colons must have them doubled (or the list-separator changed).
1586 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1587 view the policy declared in the DMARC record. Currently, $dmarc_status
1588 is a combined value of both the record presence and the result of the
1591 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1593 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1594 dnslookup router and the smtp transport (applying to the forward
1597 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1598 of ldap servers used for a specific lookup. Patch provided by Heiko
1601 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1602 New variable $lookup_dnssec_authenticated for observability.
1604 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1605 Patch submitted by Lars Timman.
1607 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1609 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1610 Requires trusted mode and valid format message id, aborts otherwise.
1611 Patch contributed by Heiko Schlichting.
1613 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1614 certextract with support for various fields. Bug 1358.
1616 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1617 is requested by default, modifiable by smtp transport option
1620 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1621 operate on certificate variables to give certificate fingerprints
1622 Also new ${sha256:cert_variable}.
1624 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1626 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1627 Christian Aistleitner.
1629 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1631 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1632 file. Patch from Wolfgang Breyha.
1634 JH/25 Expand the coverage of the delivery $host and $host_address to
1635 client authenticators run in verify callout. Bug 1476.
1637 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1638 align with daemon_smtp_ports. Bug 72.
1640 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1641 support and error reporting did not work properly.
1643 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1644 and is readable. Patch from Andrew Colin Kissa.
1646 TL/14 Enhance documentation of ${run expansion and how it parses the
1647 commandline after expansion, particularly in the case when an
1648 unquoted variable expansion results in an empty value.
1650 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1652 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1653 by renaming away from it. Addresses GH issue 15, reported by
1656 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1658 TL/15 SECURITY: prevent double expansion in math comparison functions
1659 (can expand unsanitized data). Not remotely exploitable.
1666 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1668 PP/02 Make -n do something, by making it not do something.
1669 When combined with -bP, the name of an option is not output.
1671 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1674 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1675 $sender_host_name and config options to manage this, and basic check
1678 PP/05 DSCP support for outbound connections and control modifier for inbound.
1680 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1681 (Only plugin which currently uses this is kerberos4, which nobody should
1682 be using, but we should make it available and other future plugins might
1683 conceivably use it, even though it would break NAT; stuff *should* be
1684 using channel bindings instead).
1686 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1687 name; added for Sendmail compatibility; requires admin caller.
1688 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1689 just ignore it); requires trusted caller.
1690 Also parse but ignore: -Ac -Am -X<logfile>
1693 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1695 TL/02 Add +smtp_confirmation as a default logging option.
1697 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1698 Patch by Magnus Holmgren from 2007-02-20.
1700 TL/04 Bugzilla 1281 - Spec typo.
1701 Bugzilla 1283 - Spec typo.
1702 Bugzilla 1290 - Spec grammar fixes.
1704 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1706 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1708 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1709 mailing list by by Dmitry Isaikin.
1711 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1713 JH/02 Support "G" suffix to numbers in ${if comparisons.
1715 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1717 NM/01 Bugzilla 1197 - Spec typo
1718 Bugzilla 1196 - Spec examples corrections
1720 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1722 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1723 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1726 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1727 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1729 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1730 "acl {{name}{arg}...}", and optional args on acl condition
1733 JH/05 Permit multiple router/transport headers_add/remove lines.
1735 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1737 JH/07 Avoid using a waiting database for a single-message-only transport.
1738 Performance patch from Paul Fisher. Bugzilla 1262.
1740 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1743 JH/09 Add $headers_added variable, with content from use of ACL modifier
1744 add_header (but not yet added to the message). Bugzilla 199.
1746 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1747 Pulled from Bugzilla 817 by Wolfgang Breyha.
1749 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1751 (nb: this is the same fix as in Exim 4.80.1)
1753 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1756 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1759 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1761 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1762 advertise SMTP AUTH mechanism to us, instead of a generic
1763 protocol violation error. Also, make Exim more robust to bad
1764 data from the Dovecot auth socket.
1766 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1768 When a queue runner is handling a message, Exim first routes the
1769 recipient addresses, during which it prunes them based on the retry
1770 hints database. After that it attempts to deliver the message to
1771 any remaining recipients. It then updates the hints database using
1774 So if a recipient address works intermittently, it can get repeatedly
1775 deferred at routing time. The retry hints record remains fresh so the
1776 address never reaches the final cutoff time.
1778 This is a fairly common occurrence when a user is bumping up against
1779 their storage quota. Exim had some logic in its local delivery code
1780 to deal with this. However it did not apply to per-recipient defers
1781 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1783 This change adds a proper retry rule check during routing so that the
1784 final cutoff time is checked against the message's age. We only do
1785 this check if there is an address retry record and there is not a
1786 domain retry record; this implies that previous attempts to handle
1787 the address had the retry_use_local_parts option turned on. We use
1788 this as an approximation for the destination being like a local
1789 delivery, as in LMTP.
1791 I suspect this new check makes the old local delivery cutoff check
1792 redundant, but I have not verified this so I left the code in place.
1794 TF/02 Correct gecos expansion when From: is a prefix of the username.
1796 Test 0254 submits a message to Exim with the header
1800 When I ran the test suite under the user fanf2, Exim expanded
1801 the header to contain my full name, whereas it should have added
1802 a Resent-Sender: header. It erroneously treats any prefix of the
1803 username as equal to the username.
1805 This change corrects that bug.
1807 GF/01 DCC debug and logging tidyup
1808 Error conditions log to paniclog rather than rejectlog.
1809 Debug lines prefixed by "DCC: " to remove any ambiguity.
1811 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1813 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1814 Bug spotted by Jeremy Harris; was flawed since initial commit.
1815 Would have resulted in OCSP responses post-SNI triggering an Exim
1816 NULL dereference and crash.
1818 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1820 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1821 Bug detection, analysis and fix by Samuel Thibault.
1822 Bugzilla 1331, Debian bug #698092.
1824 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1826 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1827 Server implementation by Todd Lyons, client by JH.
1828 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1829 config variable "prdr_enable" controls whether the server
1830 advertises the facility. If the client requests PRDR a new
1831 acl_data_smtp_prdr ACL is called once for each recipient, after
1832 the body content is received and before the acl_smtp_data ACL.
1833 The client is controlled by both of: a hosts_try_prdr option
1834 on the smtp transport, and the server advertisement.
1835 Default client logging of deliveries and rejections involving
1836 PRDR are flagged with the string "PRDR".
1838 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1839 fclose(). Diagnosis by Todd Lyons.
1841 PP/17 Update configure.default to handle IPv6 localhost better.
1842 Patch by Alain Williams (plus minor tweaks).
1845 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1846 This is now consistent with GnuTLS, and is now documented: the
1847 previous undocumented portable approach to treating the option as
1848 unset was to force an expansion failure. That still works, and
1849 an empty string is now equivalent.
1851 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1852 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1853 not performing validation itself.
1855 PP/20 Added force_command boolean option to pipe transport.
1856 Patch from Nick Koston, of cPanel Inc.
1858 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1861 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1863 PP/21 Fix eximon continuous updating with timestamped log-files.
1864 Broken in a format-string cleanup in 4.80, missed when I repaired the
1865 other false fix of the same issue.
1866 Report and fix from Heiko Schlichting.
1869 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1870 Report from Prashanth Katuri.
1872 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1873 It's SecureTransport, so affects any MacOS clients which use the
1874 system-integrated TLS libraries, including email clients.
1876 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1877 using a MIME ACL for non-SMTP local injection.
1878 Report and assistance in diagnosis by Warren Baker.
1880 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1882 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1884 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1885 last id that failed so it may be referenced in subsequent ACL's.
1887 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1890 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1891 ldap library initialization, allowing self-signed CA's to be
1892 used. Also properly sets require_cert option later in code by
1893 using NULL (global ldap config) instead of ldap handle (per
1894 session). Bug diagnosis and testing by alxgomz.
1896 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1897 the src/util/ subdirectory.
1899 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1900 renamed to Transport Post Delivery Action by Jeremy Harris, as
1903 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1904 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1905 redis_servers = needs to be configured which will be used by the redis
1906 lookup. Patch from Warren Baker, of The Packet Hub.
1908 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1910 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1911 hostname or reverse DNS when processing a host list. Used suggestions
1912 from multiple comments on this bug.
1914 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1916 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1917 Missed a few lines, added it to make the runtest require no keyboard
1920 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1921 contains upper case chars. Make router use caseful_local_part.
1923 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1924 support when GnuTLS has been built with p11-kit.
1930 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1932 This, or similar/improved, will also be change PP/11 of 4.82.
1938 PP/01 Handle short writes when writing local log-files.
1939 In practice, only affects FreeBSD (8 onwards).
1940 Bugzilla 1053, with thanks to Dmitry Isaikin.
1942 NM/01 Bugzilla 949 - Documentation tweak
1944 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1947 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1949 PP/02 Implemented gsasl authenticator.
1951 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1953 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1954 `pkg-config foo` for cflags/libs.
1956 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1957 with rest of GSASL and with heimdal_gssapi.
1959 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1960 `pkg-config foo` for cflags/libs for the TLS implementation.
1962 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1963 properties get this fed in as external SSF. A number of robustness
1964 and debugging improvements to the cyrus_sasl authenticator.
1966 PP/08 cyrus_sasl server now expands the server_realm option.
1968 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1969 Patch by Jeremy Harris.
1971 PP/10 Added dbmjz lookup type.
1973 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1975 PP/12 MAIL args handles TAB as well as SP, for better interop with
1976 non-compliant senders.
1977 Analysis and variant patch by Todd Lyons.
1979 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1980 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1981 Patch from Dirk Mueller <dmueller@suse.com>
1983 PP/13 tls_peerdn now print-escaped for spool files.
1984 Observed some $tls_peerdn in wild which contained \n, which resulted
1985 in spool file corruption.
1987 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1988 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1989 or write after TLS renegotiation, which otherwise led to messages
1992 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1993 as a tracking header (ie: a signed header comes before the signature).
1994 Patch from Wolfgang Breyha.
1996 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1997 comma-sep list; embedded commas doubled.
1999 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
2001 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
2003 Report and patch from Dmitry Banschikov.
2005 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
2006 Removed SSL_clear() after SSL_new() which led to protocol negotiation
2007 failures. We appear to now support TLS1.1+ with Exim.
2009 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
2010 lets Exim select keys and certificates based upon TLS SNI from client.
2011 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
2012 before an outbound SMTP session. New log_selector, +tls_sni.
2014 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
2015 NULL dereference. Report and patch from Alun Jones.
2017 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
2018 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
2019 Not seeing resolver debug output on NetBSD, but suspect this is a
2020 resolver implementation change.
2022 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
2023 Left warnings. Added "eximon gdb" invocation mode.
2025 PP/21 Defaulting "accept_8bitmime" to true, not false.
2027 PP/22 Added -bw for inetd wait mode support.
2029 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
2030 locate the relevant includes and libraries. Made this the default.
2032 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
2033 Bugzilla 1246, report and most of solution from Tomasz Kusy.
2035 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
2036 This may cause build issues on older platforms.
2038 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
2039 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
2040 gnutls_require_mac & gnutls_require_protocols (no longer supported).
2041 Added SNI support via GnuTLS too.
2042 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
2044 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
2046 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
2047 Applied second patch from Janne, implementing suggestion to default
2048 multiple-strings-in-record handling to match SPF spec.
2050 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
2052 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
2053 read-only, out of scope).
2054 Patch from Wolfgang Breyha, report from Stuart Northfield.
2056 PP/29 Fix three issues highlighted by clang analyser static analysis.
2057 Only crash-plausible issue would require the Cambridge-specific
2058 iplookup router and a misconfiguration.
2059 Report from Marcin Mirosław.
2061 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
2063 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
2064 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
2065 As part of this, removing so much warning spew let me fix some minor
2066 real issues in debug logging.
2068 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
2069 assignment on my part. Fixed.
2071 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
2072 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
2073 Janne Snabb (who went above and beyond: thank you).
2075 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
2076 string otherwise requires a connection and a bunch more work and it's
2077 relatively easy to get wrong. Should also expose TLS library linkage
2080 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
2081 64-bit ${eval} (JH/03).
2083 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
2084 GNU libc to support some of the 64-bit stuff, should not lead to
2085 conflicts. Defined before os.h is pulled in, so if a given platform
2086 needs to override this, it can.
2088 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
2089 protection layer was required, which is not implemented.
2090 Bugzilla 1254, patch from Wolfgang Breyha.
2092 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
2093 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
2094 tls_dhparam take prime identifiers. Also unbreak combination of
2095 OpenSSL+DH_params+TLSSNI.
2097 PP/39 Disable SSLv2 by default in OpenSSL support.
2103 PP/01 Solaris build fix for Oracle's LDAP libraries.
2104 Bugzilla 1109, patch from Stephen Usher.
2106 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
2108 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
2111 TF/02 Fix a couple more cases where we did not log the error message
2112 when unlink() failed. See also change 4.74-TF/03.
2114 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
2115 lock up or crash if it happened to be inside a call to libc when it
2116 got a SIGUSR1 from exiwhat.
2118 The SIGUSR1 handler appends the current process status to the process
2119 log which is later printed by exiwhat. It used to use the general
2120 purpose logging code to do this, but several functions it calls are
2121 not safe for signals.
2123 The new output code in the SIGUSR1 handler is specific to the process
2124 log, and simple enough that it's easy to inspect for signal safety.
2125 Removing some special cases also simplifies the general logging code.
2126 Removing the spurious timestamps from the process log simplifies
2129 TF/04 Improved ratelimit ACL condition.
2131 The /noupdate option has been deprecated in favour of /readonly which
2132 has clearer semantics. The /leaky, /strict, and /readonly update modes
2133 are mutually exclusive. The update mode is no longer included in the
2134 database key; it just determines when the database is updated. (This
2135 means that when you upgrade Exim will forget old rate measurements.)
2137 Exim now checks that the per_* options are used with an update mode that
2138 makes sense for the current ACL. For example, when Exim is processing a
2139 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
2140 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
2141 must specify per_mail/readonly. If you omit the update mode it defaults to
2142 /leaky where that makes sense (as before) or /readonly where required.
2144 The /noupdate option is now undocumented but still supported for
2145 backwards compatibility. It is equivalent to /readonly except that in
2146 ACLs where /readonly is required you may specify /leaky/noupdate or
2147 /strict/noupdate which are treated the same as /readonly.
2149 A useful new feature is the /count= option. This is a generalization
2150 of the per_byte option, so that you can measure the throughput of other
2151 aggregate values. For example, the per_byte option is now equivalent
2152 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
2154 The per_rcpt option has been generalized using the /count= mechanism
2155 (though it's more complicated than the per_byte equivalence). When it is
2156 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
2157 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
2158 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
2159 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
2160 Note that using per_rcpt with a non-readonly update mode in more than
2161 one ACL will cause the recipients to be double-counted. (The per_mail
2162 and per_byte options don't have this problem.)
2164 The handling of very low rates has changed slightly. If the computed rate
2165 is less than the event's count (usually one) then this event is the first
2166 after a long gap. In this case the rate is set to the same as this event's
2167 count, so that the first message of a spam run is counted properly.
2169 The major new feature is a mechanism for counting the rate of unique
2170 events. The new per_addr option counts the number of different
2171 recipients that someone has sent messages to in the last time period. It
2172 behaves like per_rcpt if all the recipient addresses are different, but
2173 duplicate recipient addresses do not increase the measured rate. Like
2174 the /count= option this is a general mechanism, so the per_addr option
2175 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
2176 example, measure the rate that a client uses different sender addresses
2177 with the options per_mail/unique=$sender_address. There are further
2178 details in the main documentation.
2180 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
2182 TF/06 Removed a few PCRE remnants.
2184 TF/07 Automatically extract Exim's version number from tags in the git
2185 repository when doing development or release builds.
2187 PP/02 Raise smtp_cmd_buffer_size to 16kB.
2188 Bugzilla 879. Patch from Paul Fisher.
2190 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
2191 Heavily based on revision 40f9a89a from Simon Arlott's tree.
2194 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
2196 PP/05 Variable $av_failed, true if the AV scanner deferred.
2197 Bugzilla 1078. Patch from John Horne.
2199 PP/06 Stop make process more reliably on build failure.
2200 Bugzilla 1087. Patch from Heiko Schlittermann.
2202 PP/07 Make maildir_use_size_file an _expandable_ boolean.
2203 Bugzilla 1089. Patch from Heiko Schlittermann.
2205 PP/08 Handle ${run} returning more data than OS pipe buffer size.
2206 Bugzilla 1131. Patch from Holger Weiß.
2208 PP/09 Handle IPv6 addresses with SPF.
2209 Bugzilla 860. Patch from Wolfgang Breyha.
2211 PP/10 GnuTLS: support TLS 1.2 & 1.1.
2213 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
2216 PP/11 match_* no longer expand right-hand-side by default.
2217 New compile-time build option, EXPAND_LISTMATCH_RHS.
2218 New expansion conditions, "inlist", "inlisti".
2220 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
2222 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
2224 PP/14 fix log_write() format string regression from TF/03.
2225 Bugzilla 1152. Patch from Dmitry Isaikin.
2231 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
2233 PP/02 Harmonised TLS library version reporting; only show if debugging.
2234 Layout now matches that introduced for other libraries in 4.74 PP/03.
2236 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
2238 PP/04 New "dns_use_edns0" global option.
2240 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
2243 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
2244 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
2246 TK/01 Updated PolarSSL code to 0.14.2.
2247 Bugzilla 1097. Patch from Andreas Metzler.
2249 PP/07 Catch divide-by-zero in ${eval:...}.
2250 Fixes bugzilla 1102.
2252 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
2255 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
2256 format-string attack -- SECURITY: remote arbitrary code execution.
2258 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
2259 time unintentionally subject to list matching rules, letting the header
2260 cause arbitrary Exim lookups (of items which can occur in lists, *not*
2261 arbitrary string expansion). This allowed for information disclosure.
2263 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
2264 INT_MIN/-1 -- value coerced to INT_MAX.
2270 NM/01 Workaround for PCRE version dependency in version reporting
2273 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
2274 This fixes portability to compilers other than gcc, notably
2275 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
2277 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
2278 makefiles for portability to HP-UX and POSIX correctness.
2280 PP/01 Permit LOOKUP_foo enabling on the make command-line.
2281 Also via indented variable definition in the Makefile.
2282 (Debugging by Oliver Heesakkers).
2284 PP/02 Restore caching of spamd results with expanded spamd_address.
2285 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
2287 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
2288 Improves build reliability. Fix from: Frank Elsner
2290 NM/02 Fix wide character breakage in the rfc2047 coding
2291 Fixes bug 1064. Patch from Andrey N. Oktyabrski
2293 NM/03 Allow underscore in dnslist lookups
2294 Fixes bug 1026. Patch from Graeme Fowler
2296 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
2297 Code patches from Adam Ciarcinski of NetBSD.
2299 NM/04 Fixed exiqgrep to cope with mailq missing size issue
2302 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
2303 is logged, to avoid truncation. Patch from John Horne.
2305 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
2306 Patch from Jakob Hirsch.
2308 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
2309 SQL string expansion failure details.
2310 Patch from Andrey Oktyabrski.
2312 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
2313 Patch from Simon Arlott.
2315 PP/09 New lookups functionality failed to compile on old gcc which rejects
2316 extern declarations in function scope.
2317 Patch from Oliver Fleischmann
2319 PP/10 Use sig_atomic_t for flags set from signal handlers.
2320 Check getgroups() return and improve debugging.
2321 Fixed developed for diagnosis in bug 927 (which turned out to be
2324 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
2325 Patch from Mark Zealey.
2327 PP/12 Bugzilla 1056: Improved spamd server selection.
2328 Patch from Mark Zealey.
2330 PP/13 Bugzilla 1086: Deal with maildir quota file races.
2331 Based on patch from Heiko Schlittermann.
2333 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
2334 Patch from Uwe Doering, sign-off by Michael Haardt.
2336 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
2337 variable declaration deep within a block. Bug and patch from
2340 PP/15 lookups-Makefile IRIX compatibility coercion.
2342 PP/16 Make DISABLE_DKIM build knob functional.
2344 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
2345 Patch by Simon Arlott
2347 TF/03 Fix valgrind.h portability to C89 compilers that do not support
2348 variable argument macros. Our copy now differs from upstream.
2354 TF/01 Failure to get a lock on a hints database can have serious
2355 consequences so log it to the panic log.
2357 TF/02 Log LMTP confirmation messages in the same way as SMTP,
2358 controlled using the smtp_confirmation log selector.
2360 TF/03 Include the error message when we fail to unlink a spool file.
2362 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
2363 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
2364 for maintaining out-of-tree patches for some time.
2366 PP/01 Bugzilla 139: Documentation and portability issues.
2367 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
2368 Handle per-OS dynamic-module compilation flags.
2370 PP/02 Let /dev/null have normal permissions.
2371 The 4.73 fixes were a little too stringent and complained about the
2372 permissions on /dev/null. Exempt it from some checks.
2373 Reported by Andreas M. Kirchwitz.
2375 PP/03 Report version information for many libraries, including
2376 Exim version information for dynamically loaded libraries. Created
2377 version.h, now support a version extension string for distributors
2378 who patch heavily. Dynamic module ABI change.
2380 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
2381 privilege escalation vulnerability whereby the Exim run-time user
2382 can cause root to append content of the attacker's choosing to
2385 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
2388 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
2389 If dropping privileges for untrusted macros, we disabled normal logging
2390 on the basis that it would fail; for the Exim run-time user, this is not
2391 the case, and it resulted in successful deliveries going unlogged.
2392 Fixed. Reported by Andreas Metzler.
2398 PP/01 Date: & Message-Id: revert to normally being appended to a message,
2399 only prepend for the Resent-* case. Fixes regression introduced in
2400 Exim 4.70 by NM/22 for Bugzilla 607.
2402 PP/02 Include check_rfc2047_length in configure.default because we're seeing
2403 increasing numbers of administrators be bitten by this.
2405 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
2407 PP/03 Bugzilla 994: added openssl_options main configuration option.
2409 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
2411 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
2413 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
2415 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
2416 without a peer certificate, leading to a segfault because of an
2417 assumption that peers always have certificates. Be a little more
2418 paranoid. Problem reported by Martin Tscholak.
2420 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
2421 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
2422 NB: ClamAV planning to remove STREAM in "middle of 2010".
2423 CL also introduces -bmalware, various -d+acl logging additions and
2424 more caution in buffer sizes.
2426 PP/09 Implemented reverse_ip expansion operator.
2428 PP/10 Bugzilla 937: provide a "debug" ACL control.
2430 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
2432 PP/12 Bugzilla 973: Implement --version.
2434 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
2436 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
2438 PP/15 Bugzilla 816: support multiple condition rules on Routers.
2440 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
2441 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
2442 ignore trailing whitespace.
2444 JJ/02 prevent non-panic DKIM error from being sent to paniclog
2446 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
2449 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
2450 Notification from Dr Andrew Aitchison.
2452 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
2453 ExtendedDetectionInfo response format.
2454 Notification from John Horne.
2456 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
2459 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
2460 XSL and documented dependency on system catalogs, with examples of how
2463 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
2466 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
2467 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
2468 configuration file which is writeable by the Exim user or group.
2470 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
2471 of configuration files to cover files specified with the -C option if
2472 they are going to be used with root privileges, not just the default
2475 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
2476 option (effectively making it always true).
2478 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
2479 files to be used while preserving root privileges.
2481 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
2482 that rogue child processes cannot use them.
2484 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
2485 run-time user, instead of root.
2487 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
2488 Exim run-time user without dropping privileges.
2490 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
2491 result string, instead of calling string_vformat() twice with the same
2494 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
2495 for other users. Others should always drop root privileges if they use
2496 -C on the command line, even for a whitelisted configure file.
2498 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
2500 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2506 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2507 $data_path, and $header_path variables; fixed documentation bugs and
2510 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2511 exipick to access non-standard spools, including the "frozen" queue
2514 NM/01 Bugzilla 965: Support mysql stored procedures.
2515 Patch from Alain Williams
2517 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2519 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2520 Patch from Andreas Metzler
2522 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2523 Patch from Kirill Miazine
2525 NM/05 Bugzilla 671: Added umask to procmail example.
2527 JJ/03 installed exipick 20100323.0, fixing doc bug
2529 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2530 directory. Notification and patch from Dan Rosenberg.
2532 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2534 TK/02 Improve log output when DKIM signing operation fails.
2536 MH/01 Treat the transport option dkim_domain as a colon separated
2537 list, not as a single string, and sign the message with each element,
2538 omitting multiple occurences of the same signer.
2540 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2541 Bugzilla 985, 986. Patch by Simon Arlott
2543 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2544 Patch by Simon Arlott
2546 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2547 MBX locking. Notification from Dan Rosenberg.
2553 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2555 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2557 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2559 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2561 NM/04 Bugzilla 915: Flag broken perl installation during build.
2567 TK/01 Added patch by Johannes Berg that expands the main option
2568 "spamd_address" if it starts with a dollar sign.
2570 TK/02 Write list of recipients to X-Envelope-Sender header when building
2571 the mbox-format spool file for content scanning (suggested by Jakob
2574 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2575 (http://www.dcc-servers.net/) support via dccifd. Activated by
2576 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2578 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2579 by Mark Daniel Reidel <mr@df.eu>.
2581 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2582 When building exim an external PCRE library is now needed -
2583 PCRE is a system library on the majority of modern systems.
2584 See entry on PCRE_LIBS in EDITME file.
2586 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2587 conversation. Added nologin parameter to request.
2588 Patch contributed by Kirill Miazine.
2590 TF/01 Do not log submission mode rewrites if they do not change the address.
2592 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2594 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2595 log files in place. Contributed by Roberto Lima.
2597 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2599 TF/03 Bugzilla 615: When checking the local_parts router precondition
2600 after a local_part_suffix or local_part_prefix option, Exim now
2601 does not use the address's named list lookup cache, since this
2602 contains cached lookups for the whole local part.
2604 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2605 Robert Millan. Documentation is in experimental-spec.txt.
2607 TF/04 Bugzilla 668: Fix parallel build (make -j).
2609 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2611 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2612 Patch provided by Jan Srzednicki.
2614 TF/05 Leading white space used to be stripped from $spam_report which
2615 wrecked the formatting. Now it is preserved.
2617 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2618 that they are available at delivery time.
2620 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2622 TF/08 TLS error reporting now respects the incoming_interface and
2623 incoming_port log selectors.
2625 TF/09 Produce a more useful error message if an SMTP transport's hosts
2626 setting expands to an empty string.
2628 NM/06 Bugzilla 744: EXPN did not work under TLS.
2629 Patch provided by Phil Pennock.
2631 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2632 Patch provided by Richard Godbee.
2634 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2635 acl_smtp_notquit, added index entry.
2637 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2638 Patch provided by Eugene Bujak.
2640 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2641 accept(). Patch provided by Maxim Dounin.
2643 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2644 Patch provided by Phil Pennock.
2646 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2648 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2649 Patch provided by Brad "anomie" Jorsch.
2651 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2652 Patch provided by Dean Brooks.
2654 TK/05 Add native DKIM support (does not depend on external libraries).
2656 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2657 Patch provided by Graeme Fowler.
2659 NM/16 Bugzilla 851: Documentation example syntax fix.
2661 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2663 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2666 NM/19 Bugzilla 745: TLS version reporting.
2667 Patch provided by Phil Pennock.
2669 NM/20 Bugzilla 167: bool: condition support.
2670 Patch provided by Phil Pennock.
2672 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2673 clients. Patch provided by Phil Pennock.
2675 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2676 Patch provided by Brad "anomie" Jorsch.
2678 NM/23 Bugzilla 687: Fix misparses in eximstats.
2679 Patch provided by Heiko Schlittermann.
2681 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2682 Patch provided by Heiko Schlittermann.
2684 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2685 plus update to original patch.
2687 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2689 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2690 Patch provided by David Brownlee.
2692 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2694 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2696 NM/30 Bugzilla 888: TLS documentation bugfixes.
2698 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2700 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2701 Unlike the original bugzilla I have changed all shell scripts in src tree.
2703 NM/33 Bugzilla 898: Transport filter timeout fix.
2704 Patch by Todd Rinaldo.
2706 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2707 Patch by Serge Demonchaux.
2709 NM/35 Bugzilla 39: Base64 decode bug fixes.
2710 Patch by Jakob Hirsch.
2712 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2714 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2716 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2718 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2724 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2725 ALT-N's libdkim that I have put here:
2726 http://duncanthrax.net/exim-experimental/
2728 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2729 were called 'true' and it seems that C99 defines that as a reserved
2730 keyword to be used with 'bool' variable types. That means you could
2731 not include C99-style headers which use bools without triggering
2732 build errors in sieve.c.
2734 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2735 as mailq or other aliases. Changed the --help handling significantly
2736 to do whats expected. exim_usage() emits usage/help information.
2738 SC/01 Added the -bylocaldomain option to eximstats.
2740 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2742 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2744 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2750 PH/01 Another patch from the Sieve maintainer.
2752 PH/02 When an IPv6 address is converted to a string for single-key lookup
2753 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2754 dots are used instead of colons so that keys in lsearch files need not
2755 contain colons. This was done some time before quoting was made available
2756 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2757 (notated using the quote facility) so as to distinguish them from IPv4
2758 keys. This meant that lookups for IP addresses in host lists did not work
2759 for iplsearch lookups.
2761 This has been fixed by arranging for IPv6 addresses to be expressed with
2762 colons if the lookup type is iplsearch. This is not incompatible, because
2763 previously such lookups could never work.
2765 The situation is now rather anomalous, since one *can* have colons in
2766 ordinary lsearch keys. However, making the change in all cases is
2767 incompatible and would probably break a number of configurations.
2769 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2772 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2773 conversion specification without a maximum field width, thereby enabling
2774 a rogue spamd server to cause a buffer overflow. While nobody in their
2775 right mind would setup Exim to query an untrusted spamd server, an
2776 attacker that gains access to a server running spamd could potentially
2777 exploit this vulnerability to run arbitrary code as the Exim user.
2779 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2780 $primary_hostname instead of what libspf2 thinks the hosts name is.
2782 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2783 a directory entry by the name of the lookup key. Previously, if a
2784 symlink pointed to a non-existing file or a file in a directory that
2785 Exim lacked permissions to read, a lookup for a key matching that
2786 symlink would fail. Now it is enough that a matching directory entry
2787 exists, symlink or not. (Bugzilla 503.)
2789 PH/03 The body_linecount and body_zerocount variables are now exported in the
2792 PH/04 Added the $dnslist_matched variable.
2794 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2795 This means they are set thereafter only if the connection becomes
2798 PH/06 Added the client_condition to authenticators so that some can be skipped
2799 by clients under certain conditions.
2801 PH/07 The error message for a badly-placed control=no_multiline_responses left
2802 "_responses" off the end of the name.
2804 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2806 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2807 (without spaces) instead of just copying the configuration text.
2809 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2811 PH/11 Added $max_received_linelength.
2813 PH/12 Added +ignore_defer and +include_defer to host lists.
2815 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2816 way in which PCRE > 7.0 is built.
2818 PH/14 Implemented queue_only_load_latch.
2820 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2821 MAIL command. The effect was to mangle the value on 64-bit systems.
2823 PH/16 Another patch from the Sieve maintainer.
2825 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2827 PH/18 If a system quota error occurred while trying to create the file for
2828 a maildir delivery, the message "Mailbox is full" was not appended to the
2829 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2830 only to a quota excession during the actual writing of the file.
2832 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2833 characters?) which causes problems in log lines. The DN values are now
2834 passed through string_printing() before being added to log lines.
2836 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2837 and InterBase are left for another time.)
2839 PH/21 Added message_body_newlines option.
2841 PH/22 Guard against possible overflow in moan_check_errorcopy().
2843 PH/23 POSIX allows open() to be a macro; guard against that.
2845 PH/24 If the recipient of an error message contained an @ in the local part
2846 (suitably quoted, of course), incorrect values were put in $domain and
2847 $local_part during the evaluation of errors_copy.
2853 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2854 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2857 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2858 issue a MAIL command.
2860 PH/02 In an ACL statement such as
2862 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2864 if a client was not listed at all, or was listed with a value other than
2865 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2866 the condition was not true (as it should be), so access was not denied.
2867 The bug was that the ! inversion was incorrectly passed on to the second
2868 item. This has been fixed.
2870 PH/03 Added additional dnslists conditions == and =& which are different from
2871 = and & when the dns lookup returns more than one IP address.
2873 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2874 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2876 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2877 FSYNC, which compiles an option called disable_fsync that allows for
2878 bypassing fsync(). The documentation is heavily laced with warnings.
2880 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2882 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2883 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2884 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2885 including adding "make clean"; (3) Added -fPIC when compiling the test
2886 dynamically loaded module, to get rid of a warning.
2888 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2889 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2890 = 0s. The bug is otherwise harmless.
2892 PH/07 There was a bug in the dovecot authenticator such that the value of
2893 $auth1 could be overwritten, and so not correctly preserved, after a
2894 successful authentication. This usually meant that the value preserved by
2895 the server_setid option was incorrect.
2897 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2899 PH/09 Installed PCRE release 7.0.
2901 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2902 run for batched SMTP input. It is now run at the start of every message
2903 in the batch. While fixing this I discovered that the process information
2904 (output by running exiwhat) was not always getting set for -bs and -bS
2905 input. This is fixed, and it now also says "batched" for BSMTP.
2907 PH/11 Added control=no_pipelining.
2909 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2910 patch, slightly modified), and move the expansion of helo_data till after
2911 the connection is made in the smtp transport (so it can use these
2914 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2916 PH/14 Added log_selector = +pid.
2918 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2920 PH/16 Add ${if forany and ${if forall.
2922 PH/17 Added dsn_from option to vary the From: line in DSNs.
2924 PH/18 Flush SMTP output before performing a callout, unless control =
2925 no_callout_flush is set.
2927 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2928 was true (the default) a successful delivery failed to delete the retry
2929 item, thus causing premature timeout of the address. The bug is now
2932 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2934 PH/21 Long custom messages for fakedefer and fakereject are now split up
2935 into multiline responses in the same way that messages for "deny" and
2936 other ACL rejections are.
2938 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2939 with slight modification.
2941 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2942 draft, changing the syntax and factoring some duplicate code".
2944 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2945 for deliveries of the second and subsequent messages over the same SMTP
2948 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2949 ${reduce, with only minor "tidies".
2951 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2953 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2954 expansion side effects.
2956 PH/27 When a message times out after an over-quota error from an Exim-imposed
2957 quota, the bounce message says "mailbox is full". This message was not
2958 being given when it was a system quota that was exceeded. It now should
2961 MH/03 Made $recipients available in local_scan(). local_scan() already has
2962 better access to the recipient list through recipients_list[], but
2963 $recipients can be useful in postmaster-provided expansion strings.
2965 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2966 in the case of a MAIL command with additional options following the
2967 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2968 were accidentally chopped off.
2970 PH/29 SMTP synchronization checks are implemented when a command is read -
2971 there is a check that no more input is waiting when there shouldn't be
2972 any. However, for some commands, a delay in an ACL can mean that it is
2973 some time before the response is written. In this time, more input might
2974 arrive, invalidly. So now there are extra checks after an ACL has run for
2975 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2976 pipelining has not been advertised.
2978 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2980 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2981 returned in the SMTP response when smtp_return_error_details was false.
2982 This has been fixed.
2984 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2985 instead of the C I/O that was originally supplied, because problems were
2986 reported on Solaris.
2988 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2989 Exim which did not show up earlier: it was assuming that a call to
2990 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2991 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2992 was a macro that became an assignment, so it seemed to work. This has
2993 changed to a proper function call with a void return, hence the compile
2994 error. Exim's code has been fixed.
2996 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2999 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
3001 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
3003 PH/37 If a message is not accepted after it has had an id assigned (e.g.
3004 because it turns out to be too big or there is a timeout) there is no
3005 "Completed" line in the log. When some messages of this type were
3006 selected by exigrep, they were listed as "not completed". Others were
3007 picked up by some special patterns. I have improved the selection
3008 criteria to be more general.
3010 PH/38 The host_find_failed option in the manualroute router can now be set
3011 to "ignore", to completely ignore a host whose IP address cannot be
3012 found. If all hosts are ignored, the behaviour is controlled by the new
3013 host_all_ignored option.
3015 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
3016 homing or because of multiple MX records with /mx) generated more than
3017 one IP address, and the following item turned out to be the local host,
3018 all the secondary addresses of the first item were incorrectly removed
3019 from the list, along with the local host and any following hosts (which
3020 is what is supposed to happen).
3022 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
3023 whoever called Exim into the -H file. In the case of the daemon it was
3024 behaving confusingly. When first started, it used values for whoever
3025 started the daemon, but after a SIGHUP it used the Exim user (because it
3026 calls itself on a restart). I have changed the code so that it now always
3029 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
3030 message are rejected with the same error (e.g. no authentication or bad
3031 sender address), and a DATA command is nevertheless sent (as can happen
3032 with PIPELINING or a stupid MUA), the error message that was given to the
3033 RCPT commands is included in the rejection of the DATA command. This is
3034 intended to be helpful for MUAs that show only the final error to their
3037 PH/42 Another patch from the Sieve maintainer.
3039 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
3040 Eximstats - Fixed some broken HTML links and added missing column headers
3042 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
3043 columns for Rejects, Temp Rejects, Ham, and Spam rows.
3045 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
3047 PH/43 Yet another patch from the Sieve maintainer.
3049 PH/44 I found a way to check for a TCP/IP connection going away before sending
3050 the response to the final '.' that terminates a message, but only in the
3051 case where the client has not sent further data following the '.'
3052 (unfortunately, this is allowed). However, in many cases there won't be
3053 any further data because there won't be any more messages to send. A call
3054 to select() can be used: if it shows that the input is "ready", there is
3055 either input waiting, or the socket has been closed. An attempt to read
3056 the next input character can distinguish the two cases. Previously, Exim
3057 would have sent an OK response which the client would never have see.
3058 This could lead to message repetition. This fix should cure that, at
3059 least in a lot of common cases.
3061 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
3062 advertised in response to EHLO.
3068 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
3069 fixed by 4.65/MH/01 (is this a record?) are fixed:
3071 (i) An empty string was always treated as zero by the numeric comparison
3072 operators. This behaviour has been restored.
3074 (ii) It is documented that the numeric comparison operators always treat
3075 their arguments as decimal numbers. This was broken in that numbers
3076 starting with 0 were being interpreted as octal.
3078 While fixing these problems I realized that there was another issue that
3079 hadn't been noticed. Values of message_size_limit (both the global option
3080 and the transport option) were treated as octal if they started with 0.
3081 The documentation was vague. These values are now always treated as
3082 decimal, and I will make that clear in the documentation.
3088 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
3089 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
3092 MH/01 Don't check that the operands of numeric comparison operators are
3093 integers when their expansion is in "skipping" mode (fixes bug
3094 introduced by 4.64-PH/07).
3096 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
3097 child addresses, Exim now panics and dies. Previously, because the count
3098 is held in a short int, deliveries were likely to be lost. As such a
3099 large number of recipients for a single message is ridiculous
3100 (performance will be very, very poor), I have chosen to impose a limit
3101 rather than extend the field.
3107 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
3108 leftover -K file (the existence of which was triggered by #402).
3109 While we were at it, introduced process PID as part of the -K
3110 filename. This should rule out race conditions when creating
3113 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
3114 processing considerably. Previous code took too long for large mails,
3115 triggering a timeout which in turn triggers #401.
3117 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
3118 in the DK code in transports.c. sendfile() is not really portable,
3119 hence the _LINUX specificness.
3121 TF/01 In the add_headers option to the mail command in an Exim filter,
3122 there was a bug that Exim would claim a syntax error in any
3123 header after the first one which had an odd number of characters
3126 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
3127 callout verification, Exim cached a "reject" for the entire domain. This
3128 is correct for most verifications, but it is not correct for a recipient
3129 verification with use_sender or use_postmaster set, because in that case
3130 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
3131 case of MAIL FROM:<> rejection from other early rejections (e.g.
3132 rejection of HELO). When verifying a recipient using a non-null MAIL
3133 address, the cache is ignored if it shows MAIL FROM:<> rejection.
3134 Whatever the result of the callout, the value of the domain cache is
3135 left unchanged (for any other kind of callout, getting as far as trying
3136 RCPT means that the domain itself is ok).
3138 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
3141 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
3142 manifest itself as EPIPE rather than ECONNECT. When tidying away a
3143 session, the daemon ignores ECONNECT errors and logs others; it now
3144 ignores EPIPE as well.
3146 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
3147 (quoted-printable decoding).
3149 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
3150 later the small subsequent patch to fix an introduced bug.
3152 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
3154 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
3156 PH/08 An error is now given if message_size_limit is specified negative.
3158 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
3159 to be given (somewhat) arbitrary names.
3161 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
3164 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
3165 miscellaneous code fixes
3167 PH/10 Added the log_reject_target ACL modifier to specify where to log
3170 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
3171 hostname. This is wrong, because it relates to the incoming message (and
3172 probably the interface on which it is arriving) and not to the outgoing
3173 callout (which could be using a different interface). This has been
3174 changed to use the value of the helo_data option from the smtp transport
3175 instead - this is what is used when a message is actually being sent. If
3176 there is no remote transport (possible with a router that sets up host
3177 addresses), $smtp_active_hostname is used.
3179 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
3180 tweaks were necessary in order to get it to work (see also 21 below):
3181 (a) The code assumed that strncpy() returns a negative number on buffer
3182 overflow, which isn't the case. Replaced with Exim's string_format()
3184 (b) There were several signed/unsigned issues. I just did the minimum
3185 hacking in of casts. There is scope for a larger refactoring.
3186 (c) The code used strcasecmp() which is not a standard C function.
3187 Replaced with Exim's strcmpic() function.
3188 (d) The code set only $1; it now sets $auth1 as well.
3189 (e) A simple test gave the error "authentication client didn't specify
3190 service in request". It would seem that Dovecot has changed its
3191 interface. Fortunately there's a specification; I followed it and
3192 changed what the client sends and it appears to be working now.
3194 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
3197 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
3198 address A is aliased to B and C, where B exists and C does not. Without
3199 -v the output is "A verified" because verification stops after a
3200 successful redirection if more than one address is generated. However,
3201 with -v the child addresses are also verified. Exim was outputting "A
3202 failed to verify" and then showing the successful verification for C,
3203 with its parentage. It now outputs "B failed to verify", showing B's
3204 parentage before showing the successful verification of C.
3206 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
3207 look up a TXT record in a specific list after matching in a combined
3210 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
3211 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
3212 they consult the DNS. I had assumed they would set it the way they
3213 wanted; and indeed my experiments on Linux seem to show that in some
3214 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
3215 To be on the safe side, however, I have now made the interface to
3216 host_find_byname() similar to host_find_bydns(), with an argument
3217 containing the DNS resolver options. The host_find_byname() function now
3218 sets these options at its start, just as host_find_bydns() does. The smtp
3219 transport options dns_qualify_single and dns_search_parents are passed to
3220 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
3221 of host_find_byname() use the default settings of RES_DEFNAMES
3222 (qualify_single) but not RES_DNSRCH (search_parents).
3224 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
3225 spool_read_header() do less string testing, by means of a preliminary
3226 switch on the second character of optional "-foo" lines. (This is
3227 overdue, caused by the large number of possibilities that now exist.
3228 Originally there were few.) While I was there, I also converted the
3229 str(n)cmp tests so they don't re-test the leading "-" and the first
3230 character, in the hope this might squeeze out yet more improvement.
3232 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
3233 flag allowing group syntax was set by the header_syntax check but not
3234 turned off, possible causing trouble later; (2) The flag was not being
3235 set at all for the header_verify test, causing "group"-style headers to
3236 be rejected. I have now set it in this case, and also caused header_
3237 verify to ignore an empty address taken from a group. While doing this, I
3238 came across some other cases where the code for allowing group syntax
3239 while scanning a header line wasn't quite right (mostly, not resetting
3240 the flag correctly in the right place). These bugs could have caused
3241 trouble for malformed header lines. I hope it is now all correct.
3243 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
3244 with the "reply" argument non-NULL. The code, however (which originally
3245 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
3246 but it didn't always do it. This confused somebody who was copying the
3247 code for some other use. I have removed all the tests.
3249 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
3250 feature that was used to support insecure browsers during the U.S. crypto
3251 embargo. It requires special client support, and Exim is probably the
3252 only MTA that supported it -- and would never use it because real RSA is
3253 always available. This code has been removed, because it had the bad
3254 effect of slowing Exim down by computing (never used) parameters for the
3255 RSA_EXPORT functionality.
3257 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
3258 authenticator to fail if there's a tab character in the incoming data
3259 (there should never be unless someone is messing about, as it's supposed
3260 to be base64-encoded). Also added, on Timo's advice, the "secured" option
3261 if the connection is using TLS or if the remote IP is the same as the
3262 local IP, and the "valid-client-cert option" if a client certificate has
3265 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
3266 authenticators. This can be used for authorization after authentication
3267 succeeds. (In the case of plaintext, it servers for both authentication
3270 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
3271 if any retry times were supplied.
3273 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
3274 connection, where there is no client IP address to check. In this
3275 situation, the verify now always succeeds.
3277 PH/25 Applied John Jetmore's -Mset patch.
3279 PH/26 Added -bem to be like -Mset, but loading a message from a file.
3281 PH/27 In a string expansion for a processed (not raw) header when multiple
3282 headers of the same name were present, leading whitespace was being
3283 removed from all of them, but trailing whitespace was being removed only
3284 from the last one. Now trailing whitespace is removed from each header
3285 before concatenation. Completely empty headers in a concatenation (as
3286 before) are ignored.
3288 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
3289 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
3291 PH/29 [Removed. This was a change that I later backed out, and forgot to
3292 correct the ChangeLog entry (that I had efficiently created) before
3293 committing the later change.]
3295 PH/30 Exim was sometimes attempting to deliver messages that had suffered
3296 address errors (4xx response to RCPT) over the same connection as other
3297 messages routed to the same hosts. Such deliveries are always "forced",
3298 so retry times are not inspected. This resulted in far too many retries
3299 for the affected addresses. The effect occurred only when there were more
3300 hosts than the hosts_max_try setting in the smtp transport when it had
3301 the 4xx errors. Those hosts that it had tried were not added to the list
3302 of hosts for which the message was waiting, so if all were tried, there
3303 was no problem. Two fixes have been applied:
3305 (i) If there are any address or message errors in an SMTP delivery, none
3306 of the hosts (tried or untried) are now added to the list of hosts
3307 for which the message is waiting, so the message should not be a
3308 candidate for sending over the same connection that was used for a
3309 successful delivery of some other message. This seems entirely
3310 reasonable: after all the message is NOT "waiting for some host".
3311 This is so "obvious" that I'm not sure why it wasn't done
3312 previously. Hope I haven't missed anything, but it can't do any
3313 harm, as the worst effect is to miss an optimization.
3315 (ii) If, despite (i), such a delivery is accidentally attempted, the
3316 routing retry time is respected, so at least it doesn't keep
3317 hammering the server.
3319 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
3320 in ${readsocket because some servers need this prod.
3322 PH/32 Added some extra debug output when updating a wait-xxx database.
3324 PH/33 The hint "could be header name not terminated by colon", which has been
3325 given for certain expansion errors for a long time, was not being given
3326 for the ${if def:h_colon_omitted{... case.
3328 PH/34 The spec says: "With one important exception, whenever a domain list is
3329 being scanned, $domain contains the subject domain." There was at least
3330 one case where this was not true.
3332 PH/35 The error "getsockname() failed: connection reset by peer" was being
3333 written to the panic log as well as the main log, but it isn't really
3334 panic-worthy as it just means the connection died rather early on. I have
3335 removed the panic log writing for the ECONNRESET error when getsockname()
3338 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
3339 runs only) independently of the message's sender address. This meant
3340 that, if the 4xx error was in fact related to the sender, a different
3341 message to the same recipient with a different sender could confuse
3342 things. In particular, this can happen when sending to a greylisting
3343 server, but other circumstances could also provoke similar problems.
3344 I have changed the default so that the retry time for these errors is now
3345 based a combination of the sender and recipient addresses. This change
3346 can be overridden by setting address_retry_include_sender=false in the
3349 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
3350 remote server are returned as part of bounce messages. This was not
3351 happening for LMTP over a pipe (the lmtp transport), but now it is the
3352 same for both kinds of LMTP.
3354 PH/38 Despite being documented as not happening, Exim was rewriting addresses
3355 in header lines that were in fact CNAMEs. This is no longer the case.
3357 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
3358 and queue runs started by the daemon processed all messages. This has
3359 been fixed so that -R and -S can now usefully be given with -q<time>.
3361 PH/40 Import PCRE release 6.7 (fixes some bugs).
3363 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
3365 PH/42 Give an error if -q is specified more than once.
3367 PH/43 Renamed the variables $interface_address and $interface_port as
3368 $received_ip_address and $received_port, to make it clear that these
3369 values apply to message reception, and not to the outgoing interface when
3370 a message is delivered. (The old names remain recognized, of course.)
3372 PH/44 There was no timeout on the connect() call when using a Unix domain
3373 socket in the ${readsocket expansion. There now is.
3375 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
3376 be meaningful with "accept".
3378 SC/01 Eximstats V1.43
3379 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
3381 SC/02 Eximstats V1.44
3382 Use a glob alias rather than an array ref in the generated
3383 parser. This improves both readability and performance.
3385 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
3386 Collect SpamAssassin and rejection statistics.
3387 Don't display local sender or destination tables unless
3388 there is data to show.
3389 Added average volumes into the top table text output.
3391 SC/04 Eximstats V1.46
3392 Collect data on the number of addresses (recipients)
3393 as well as the number of messages.
3395 SC/05 Eximstats V1.47
3396 Added 'Message too big' to the list of mail rejection
3397 reasons (thanks to Marco Gaiarin).
3399 SC/06 Eximstats V1.48
3400 Mainlog lines which have GMT offsets and are too short to
3401 have a flag are now skipped.
3403 SC/07 Eximstats V1.49 (Alain Williams)
3404 Added the -emptyok flag.
3406 SC/08 Eximstats V1.50
3407 Fixes for obtaining the IP address from reject messages.
3409 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
3410 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
3411 whitespace changes from 4.64-PH/27
3413 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
3416 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
3417 are found, allow negative numbers in numeric criteria)
3419 JJ/06 exipick.20061117.2, added new $message_body_missing variable
3421 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
3422 to match changes made in 4.64-PH/43
3424 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
3426 PH/47 Put in an explicit test for a DNS lookup of an address record where the
3427 "domain" is actually an IP address, and force a failure. This locks out
3428 those revolvers/nameservers that support "A-for-A" lookups, in
3429 contravention of the specifications.
3431 PH/48 When a host name was looked up from an IP address, and the subsequent
3432 forward lookup of the name timed out, the host name was left in
3433 $sender_host_name, contrary to the specification.
3435 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
3436 restricted to single-key lookups, Exim was not diagnosing an error if
3437 * or *@ was used with a query-style lookup.
3439 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
3441 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
3442 long ago, but noone interested enough thought of it. Let's just say that
3443 the "1.1" means that there are some new functions that weren't there at
3444 some point in the past.
3446 PH/51 Error processing for expansion failure of helo_data from an smtp
3447 transport during callout processing was broken.
3449 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
3450 tested/used via the -bh/-bhc/-bs options.
3452 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
3453 bug, fixed in subsequent PCRE releases).
3455 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
3456 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
3458 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
3464 SC/01 Use a glob alias rather than an array ref in eximstats generated
3465 parser. This improves both readability and performance.
3467 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
3468 Don't display local sender or destination tables in eximstats unless
3469 there is data to show.
3470 Added average volumes into the eximstats top table text output.
3472 SC/03 Collect data on the number of addresses (recipients) as well
3473 as the number of messages in eximstats.
3475 TF/01 Correct an error in the documentation for the redirect router. Exim
3476 does (usually) call initgroups() when daemonizing.
3478 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
3479 with consistent privilege compared to when running as a daemon.
3481 TF/03 Note in the spec that $authenticated_id is not set for local
3482 submissions from trusted users.
3484 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
3485 Thanks to Dean Brooks <dean@iglou.com> for the patch.
3487 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
3488 by adding some example configuration directives to the default
3489 configuration file. A little bit of work is required to uncomment the
3490 directives and define how usernames and passwords are checked, but
3491 there is now a framework to start from.
3493 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
3494 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
3495 without this. I don't know how relevant this is to other LDAP libraries.
3497 PH/02 Add the verb name to the "unknown ACL verb" error.
3499 PH/03 Magnus Holmgren's patch for filter_prepend_home.
3501 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
3503 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
3504 directory not expanded when it should be if an expanded home directory
3505 was set for the address (which is overridden by the transport).
3507 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3510 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3511 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3512 because it is too late at that time, and has no effect.
3514 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3515 security issue with \' (bugzilla #107). I could not use the
3516 PQescapeStringConn() function, because it needs a PGconn value as one of
3519 PH/08 When testing addresses using -bt, indicate those final addresses that
3520 are duplicates that would not cause an additional delivery. At least one
3521 person was confused, thinking that -bt output corresponded to deliveries.
3522 (Suppressing duplicates isn't a good idea as you lose the information
3523 about possibly different redirections that led to the duplicates.)
3525 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3526 systems where poll() doesn't work, in particular OS X.
3528 PH/10 Added more information to debugging output for retry time not reached.
3530 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3531 operations in malware.c.
3533 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3536 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3537 syslog_duplication set false, log lines that would normally be written
3538 both the the main log and to the reject log were not written to syslog at
3541 PH/14 In the default configuration, change the use of "message" in ACL warn
3542 statements to "add_header".
3544 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3545 not followed by a command (e.g. "seen endif").
3547 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3548 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3551 PH/17 Added extra conditions to the default value of delay_warning_condition
3555 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3556 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3557 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3560 The Auto-Submitted: and various List- headers are standardised, whereas I
3561 don't think Precedence: ever was.
3563 PH/18 Refactored debugging code in route_finduser() to show more information,
3564 in particular, the error code if getpwnam() issues one.
3566 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3567 This is apparently needed in addition to the PH/07 change above to avoid
3568 any possible encoding problems.
3570 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3571 but not after initializing Perl.
3573 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3574 output them only if debugging. By default they are written stderr,
3575 apparently, which is not desirable.
3577 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3580 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3583 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3585 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3586 authenticated or an ident call has been made. Suppress the default
3587 values for $authenticated_id and $authenticated_sender (but permit -oMai
3588 and -oMas) when testing with -bh.
3590 PH/24 Re-jigged the order of the tests in the default configuration so that the
3591 tests for valid domains and recipients precede the DNS black list and CSA
3592 tests, on the grounds that those ones are more expensive.
3594 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3595 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3596 This bug exists in every version of Exim that I still have, right back to
3599 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3600 However, an attempt to turn on case-sensitivity in a regex key by
3601 including (?-i) didn't work because the subject string was already
3602 lowercased, and the effects were non-intuitive. It turns out that a
3603 one-line patch can be used to allow (?-i) to work as expected.
3609 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3610 other effects) broke the use of negated acl sub-conditions.
3612 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3615 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3616 "Deny" causes Exim to reject the incoming connection with a 554 error.
3617 Unfortunately, if there is a major crisis, such as a disk failure,
3618 tcp-wrappers gives "deny", whereas what one would like would be some
3619 kind of temporary error. A kludge has been added to help with this.
3620 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3621 554 error is used if errno is still zero or contains ENOENT (which occurs
3622 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3625 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3627 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3628 errors. Otherwise a message that provokes a temporary error (when other
3629 messages do not) can cause a whole host to time out.
3631 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3632 addresses were routed directly to files or pipes from a redirect router.
3633 File deliveries just didn't batch; pipe deliveries might have suffered
3636 PH/06 A failure to get a lock for a hints database would erroneously always say
3637 "Failed to get write lock", even when it was really a read lock.
3639 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3640 of 0600. This has been changed to use the value of the lockfile_mode
3641 option (which defaults to 0600).
3643 PH/08 Applied small patch from the Sieve maintainer.
3645 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3646 folder from quota calculations, a direct delivery into this folder messed
3647 up the contents of the maildirsize file. This was because the regex was
3648 used only to exclude .Trash (or whatever) when the size of the mailbox
3649 was calculated. There was no check that a delivery was happening into an
3650 excluded directory. This bug has been fixed by ignoring all quota
3651 processing for deliveries into excluded directories.
3653 PH/10 Added the maildirfolder_create_regex option to appendfile.
3659 PH/01 The code for finding all the local interface addresses on a FreeBSD
3660 system running IPv6 was broken. This may well have applied to all BSD
3661 systems, as well as to others that have similar system calls. The broken
3662 code found IPv4 interfaces correctly, but gave incorrect values for the
3663 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3664 that it would not match correctly against @[] and not recognize the IPv6
3667 PH/02 The ipliteral router was not recognizing addresses of the form user@
3668 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3670 PH/03 Added disable_ipv6.
3672 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3673 decoded form, because it is most often used to construct To: headers
3674 lines in autoreplies, and the decoded form may well be syntactically
3675 invalid. However, $reply_address has leading white space removed, and all
3676 newlines turned into spaces so that the autoreply transport does not
3679 PH/05 If group was specified without a user on a router, and no group or user
3680 was specified on a transport, the group from the router was ignored.
3682 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3683 for visible compile-time settings that can be used to change these
3684 numbers, for those that want even more. Backwards compatibility with old
3685 spool files has been maintained. However, going back to a previous Exim
3686 release will lost any variables that are in spool files.
3688 PH/07 Two small changes when running in the test harness: increase delay when
3689 passing a TCP/IP connection to a new process, in case the original
3690 process has to generate a bounce, and remove special handling of
3691 127.0.0.2 (sic), which is no longer necessary.
3693 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3694 be the same on different OS.
3696 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3699 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3700 whether --show-vars was specified or not
3702 JJ/02 exipick: Added support for new ACL variable spool format introduced
3705 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3706 syntactically invalid From: or Reply-to: line, and a filter used this to
3707 generate an autoreply, and therefore failed to obtain an address for the
3708 autoreply, Exim could try to deliver to a non-existent relative file
3709 name, causing unrelated and misleading errors. What now happens is that
3710 it logs this as a hard delivery error, but does not attempt to create a
3713 PH/11 The exinext utility has a -C option for testing purposes, but although
3714 the given file was scanned by exinext itself; it wasn't being passed on
3715 when Exim was called.
3717 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3718 an end-of-file indication when reading a command response.
3720 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3721 compiled. In many other places in Exim, IPv6 addresses are always
3722 recognized, so I have changed this. It also means that IPv4 domain
3723 literals of the form [IPV4:n.n.n.n] are now always recognized.
3725 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3726 used if the router is not running as root, for example, when verifying at
3727 ACL time, or when using -bh. The debugging output from this situation was
3728 non-existent - all you got was a failure to exec. I have made two
3731 (a) Failures to set uid/gid, the current directory, or a process leader
3732 in a subprocess such as that created by queryprogram now generate
3733 suitable debugging output when -d is set.
3735 (b) The queryprogram router detects when it is not running as root,
3736 outputs suitable debugging information if -d is set, and then runs
3737 the subprocess without attempting to change uid/gid.
3739 PH/15 Minor change to Makefile for building test_host (undocumented testing
3742 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3743 additional section of a DNS packet that returns MX or SRV records.
3744 Instead, it always explicitly searches for A/AAAA records. This avoids
3745 major problems that occur when a DNS server includes only records of one
3746 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3747 fixed another bug: if SRV records were looked up and the corresponding
3748 address records were *not* found in the additional section, the port
3749 values from the SRV records were lost.
3751 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3752 using the correct key (the original address) when searching the retry
3753 rules in order to find which one to use for generating the retry hint.
3755 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3756 adding the default one. Similarly, if it contains a Reply-To: header, the
3757 errors_reply_to option, if set, is not used.
3759 PH/19 When calculating a retry time, Exim used to measure the "time since
3760 failure" by looking at the "first failed" field in the retry record. Now
3761 it does not use this if it is later than than the arrival time of the
3762 message. Instead it uses the arrival time. This makes for better
3763 behaviour in cases where some deliveries succeed, thus re-setting the
3764 "first failed" field. An example is a quota failure for a huge message
3765 when small messages continue to be delivered. Without this change, the
3766 "time since failure" will always be short, possible causing more frequent
3767 delivery attempts for the huge message than are intended.
3768 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3770 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3771 $1, $2, $3) because the numerical variables can be reset during some
3772 expansion items (e.g. "match"), thereby losing the authentication data.
3774 PH/21 Make -bV show the size of off_t variables so that the test suite can
3775 decide whether to run tests for quotas > 2G.
3777 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3778 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3779 filecount value is greater than 2G or if a quota value is greater than 2G
3780 on a system where the size of off_t is not greater than 4, a panic error
3783 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3784 never match. The debug and -bh output now contains an explicit error
3785 message indicating a malformed IPv4 address or mask.
3787 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3788 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3789 PH/23 above applies.
3791 PH/25 Do not write to syslog when running in the test harness. The only
3792 occasion when this arises is a failure to open the main or panic logs
3793 (for which there is an explicit test).
3795 PH/26 Added the /no_tell option to "control=freeze".
3797 PH/27 If a host name lookup failed very early in a connection, for example, if
3798 the IP address matched host_lookup and the reverse lookup yielded a name
3799 that did not have a forward lookup, an error message of the form "no IP
3800 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3801 could be logged. Now it outputs the IP address instead of "NULL".
3803 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3804 allows the sender and the authenticated sender to be set when
3805 submitting a message from within Exim. Since child_open_exim() is
3806 documented for local_scan(), the new function should be too.
3808 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3809 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3810 results in an empty string is now treated as unset.
3812 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3814 PH/31 Added sender_verify_fail logging option.
3816 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3817 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3818 tidied the source and removed it altogether.
3820 PH/33 When a queue run was abandoned because the load average was too high, a
3821 log line was always written; now it is written only if the queue_run log
3822 selector is set. In addition, the log line for abandonment now contains
3823 information about the queue run such as the pid. This is always present
3824 in "start" and "stop" lines but was omitted from the "abandon" line.
3826 PH/34 Omit spaces between a header name and the colon in the error message that
3827 is given when verify = headers_syntax fails (if there are lots of them,
3828 the message gets confusing).
3830 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3831 names, as there are now some PTR records that contain slashes. This check
3832 is only to protect against broken name servers that fall over on strange
3833 characters, so the fact that it applies to all lookups doesn't matter.
3835 PH/36 Now that the new test suite is complete, we can remove some of the
3836 special code in Exim that was needed for the old test suite. For example,
3837 sorting DNS records because real resolvers return them in an arbitrary
3838 order. The new test suite's fake resolver always returns records in the
3841 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3842 bounces) except when queue_only is set, to avoid logging races between
3843 the different processes.
3845 PH/38 Panic-die if .include specifies a non-absolute path.
3847 PH/39 A tweak to the "H" retry rule from its user.
3849 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3850 a label. They prevented compilation on older perls.
3852 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3853 a warning to be raised on newish perls.
3855 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3856 on queue. Changes to match documented behaviour of showing count of
3857 messages matching specified criteria.
3859 PH/40 Changed the default ident timeout from 30s to 5s.
3861 PH/41 Added support for the use of login_cap features, on those BSD systems
3862 that have them, for controlling the resources used by pipe deliveries.
3864 PH/42 The content-scanning code uses fopen() to create files in which to put
3865 message data. Previously it was not paying any attention to the mode of
3866 the files. Exim runs with umask(0) because the rest of the code creates
3867 files with open(), and sets the required mode explicitly. Thus, these
3868 files were ending up world-writeable. This was not a big issue, because,
3869 being within the spool directory, they were not world-accessible. I have
3870 created a function called modefopen, which takes an additional mode
3871 argument. It sets umask(777), creates the file, chmods it to the required
3872 mode, then resets the umask. All the relevant calls to fopen() in the
3873 content scanning code have been changed to use this function.
3875 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3876 to 24 hours. This avoids potential overflow problems when processing G
3877 and H retry rules. I suspect nobody ever tinkers with this value.
3879 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3881 PH/45 When the plaintext authenticator is running as a client, the server's
3882 challenges are checked to ensure they are valid base64 strings. By
3883 default, the authentication attempt is cancelled if an invalid string is
3884 received. Setting client_ignore_invalid_base64 true ignores these errors.
3885 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3886 they are received. Thus, the responses can be made to depend on the
3887 challenges. If an invalid string is ignored, an empty string is placed in
3890 PH/46 Messages that are created by the autoreply transport now contains a
3891 References: header, in accordance with RFCs 2822 and 3834.
3893 PH/47 Added authenticated_sender_force to the smtp transport.
3895 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3897 PH/49 Installed latest patch from the Sieve maintainer.
3899 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3900 was also set, the appendfile transport was unnecessarily scanning a
3901 directory of message files (e.g. for maildir delivery) to find the count
3902 of files (along with the size), even though it did not need this
3903 information. It now does the scan only if it needs to find either the
3904 size of the count of files.
3906 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3908 PH/52 Two bugs concerned with error handling when the smtp transport is
3911 (i) Exim was not creating retry information for temporary errors given
3912 for individual recipients after the DATA command when the smtp transport
3913 was used in LMTP mode. This meant that they could be retried too
3914 frequently, and not timed out correctly.
3916 (ii) Exim was setting the flag that allows error details to be returned
3917 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3918 recipients that were returned after the DATA command.
3920 PH/53 This is related to PH/52, but is more general: for any failing address,
3921 when detailed error information was permitted to be returned to the
3922 sender, but the error was temporary, then after the final timeout, only
3923 "retry timeout exceeded" was returned. Now it returns the full error as
3924 well as "retry timeout exceeded".
3926 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3927 do this, and (what is worse) MTAs that accept it.
3929 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3930 will now be deprecated.
3932 PH/56 New os.c-cygwin from the Cygwin maintainer.
3934 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3935 formats (previously only available in exim formats via -bpr, -bpru,
3936 and -bpra. Now also available in native and exiqgrep formats)
3938 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3939 with very large, slow to parse queues
3941 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3943 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3945 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3946 responses to authentication challenges, though it was showing the
3947 challenges; (ii) I've removed the CR characters from the debug output for
3950 PH/58 Allow for the insertion of a newline as well as a space when a string
3951 is turned into more than one encoded-word during RFC 2047 encoding. The
3952 Sieve code now uses this.
3954 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3955 data_4xx, lost_connection, tls_required.
3957 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3958 message was being sent as an SMTP response.
3960 PH/61 Add -l and -k options to exicyclog.
3962 PH/62 When verifying, if an address was redirected to one new address, so that
3963 verification continued, and the new address failed or deferred after
3964 having set something in $address_data, the value of $address_data was not
3965 passed back to the ACL. This was different to the case when no
3966 redirection occurred. The value is now passed back in both cases.
3968 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3969 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3970 use login_cap.h, so on its own it isn't the distinguishing feature. The
3971 new name refers directly to the setclassresources() function.
3973 PH/65 Added configuration files for NetBSD3.
3975 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3977 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3978 is preferred over IPv4.
3980 PH/68 The bounce_return_message and bounce_return_body options were not being
3981 honoured for bounces generated during the reception of non-SMTP messages.
3982 In particular, this applied to messages rejected by the ACL. This bug has
3983 been fixed. However, if bounce_return_message is true and bounce_return_
3984 body is false, the headers that are returned for a non-SMTP message
3985 include only those that have been read before the error was detected.
3986 (In the case of an ACL rejection, they have all been read.)
3988 PH/69 The HTML version of the specification is now built in a directory called
3989 spec_html instead of spec.html, because the latter looks like a path with
3990 a MIME-type, and this confuses some software.
3992 PH/70 Catch two compiler warnings in sieve.c.
3994 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3995 function verify_get_ident() calls ip_connect() to connect a socket, but
3996 if the "connect()" function timed out, ip_connect() used to close the
3997 socket. However, verify_get_ident() also closes the socket later, and in
3998 between Exim writes to the log, which may get opened at this point. When
3999 the socket was closed in ip_connect(), the log could get the same file
4000 descriptor number as the socket. This naturally causes chaos. The fix is
4001 not to close the socket in ip_connect(); the socket should be closed by
4002 the function that creates it. There was only one place in the code where
4003 this was missing, in the iplookup router, which I don't think anybody now
4004 uses, but I've fixed it anyway.
4006 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
4007 well as to direct DNS lookups. Otherwise the handling of names in host
4008 lists is inconsistent and therefore confusing.
4014 PH/01 Two changes to the default runtime configuration:
4016 (1) Move the checks for relay_from_hosts and authenticated clients from
4017 after to before the (commented out) DNS black list checks.
4019 (2) Add control=submission to the relay_from_hosts and authenticated
4020 clients checks, on the grounds that messages accepted by these
4021 statements are most likely to be submissions.
4023 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
4025 (1) Generate an error if the third argument for the ${prvs expansion is
4028 (2) Treat a missing third argument of ${prvscheck as if it were an empty
4031 (3) Reset the variables that are obtained from the first argument of
4032 ${prvscheck and used in the second argument before leaving the code,
4033 because their memory is reclaimed, so using them afterwards may do
4036 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
4037 one (it's much easier than Tom thought :-).
4039 (5) Because of (4), we can now allow for the use of $prvscheck_result
4040 inside the third argument.
4042 PH/03 For some reason, the default setting of PATH when running a command from
4043 a pipe transport was just "/usr/bin". I have changed it to
4046 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
4047 anything to be listed in the output from -bV.
4049 PH/05 When a filter generated an autoreply, the entire To: header line was
4050 quoted in the delivery log line, like this:
4052 => >A.N.Other <ano@some.domain> <original@ddress> ...
4054 This has been changed so that it extracts the operative address. There
4055 may be more than one such address. If so, they are comma-separated, like
4058 => >ano@some.domain,ona@other.domain <original@ddress> ...
4060 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
4061 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
4062 not being looked up in the rDNS to get a host name, Exim was showing the
4063 IP address twice in Received: lines, even though the IP addresses were
4064 identical. For example:
4066 Received: from [1.2.3.4] (helo=[1.2.3.4])
4068 However, if the real host name was known, it was omitting the HELO data
4069 if it matched the actual IP address. This has been tidied up so that it
4070 doesn't show the same IP address twice.
4072 PH/07 When both +timestamp and +memory debugging was on, the value given by
4073 $tod_xxx expansions could be wrong, because the tod_stamp() function was
4074 called by the debug printing, thereby overwriting the timestamp buffer.
4075 Debugging no longer uses the tod_stamp() function when +timestamp is set.
4077 PH/08 When the original message was included in an autoreply transport, it
4078 always said "this is a copy of the message, including all the headers",
4079 even if body_only or headers_only was set. It now gives an appropriate
4082 PH/09 Applied a patch from the Sieve maintainer which:
4084 o fixes some comments
4085 o adds the (disabled) notify extension core
4086 o adds some debug output for the result of if/elsif tests
4087 o points to the current vacation draft in the documentation
4088 and documents the missing references header update
4092 o fixes a bug in processing the envelope test (when testing
4093 multiple envelope elements, the last element determined the
4096 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
4097 Electronic Mail") by including:
4099 Auto-submitted: auto-generated
4101 in the messages that it generates (bounce messages and others, such as
4102 warnings). In the case of bounce messages for non-SMTP messages, there was
4103 also a typo: it was using "Auto_submitted" (underscore instead of
4104 hyphen). Since every message generated by Exim is necessarily in response
4105 to another message, thes have all been changed to:
4107 Auto-Submitted: auto-replied
4109 in accordance with these statements in the RFC:
4111 The auto-replied keyword:
4113 - SHOULD be used on messages sent in direct response to another
4114 message by an automatic process,
4116 - MUST NOT be used on manually-generated messages,
4118 - MAY be used on Delivery Status Notifications (DSNs) and Message
4119 Disposition Notifications (MDNs),
4121 - MUST NOT be used on messages generated by automatic or periodic
4122 processes, except for messages which are automatic responses to
4125 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
4126 to the default Received: header definition.
4128 PH/12 Added log selector acl_warn_skipped (default on).
4130 PH/13 After a successful wildlsearch lookup, discard the values of numeric
4131 variables because (a) they are in the wrong storage pool and (b) even if
4132 they were copied, it wouldn't work properly because of the caching.
4134 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
4135 checking when decoding. Apparently there are clients that generate
4136 overlong encoded strings. Why am I not surprised?
4138 PH/15 If the first argument of "${if match_address" was not empty, but did not
4139 contain an "@" character, Exim crashed. Now it writes a panic log message
4140 and treats the condition as false.
4142 PH/16 In autoreply, treat an empty string for "once" the same as unset.
4144 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
4145 extension "envelope-auth". The code is finished and in agreement with
4146 other implementations, but there is no documentation so far and in fact,
4147 nobody wrote the draft yet. This extension is currently #undef'ed, thus
4148 not changing the active code.
4150 Print executed "if" and "elsif" statements when debugging is used. This
4151 helps a great deal to understand what a filter does.
4153 Document more things not specified clearly in RFC3028. I had all this
4154 sorted out, when out of a sudden new issues came to my mind. Oops."
4156 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
4159 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
4160 canonical form (as documented). However, after a host name lookup from
4161 the IP address, check_host() was doing a simple string comparison with
4162 addresses acquired from the DNS when checking that the found name did
4163 have the original IP as one of its addresses. Since any found IPv6
4164 addresses are likely to be in abbreviated form, the comparison could
4165 fail. Luckily, there already exists a function for doing the comparison
4166 by converting both addresses to binary, so now that is used instead of
4167 the text comparison.
4169 PH/20 There was another similar case to PH/19, when a complete host name was
4170 given in a host list; looking up its IP address could give an abbreviated
4171 form, whereas the current host's name might or might not be abbreviated.
4172 The same fix has been applied.
4178 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
4179 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
4182 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
4183 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
4185 PH/03 Typo: missing ".o" in src/pcre/Makefile.
4187 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
4188 header line, restrict the check to what is listed in RFCs 2369 and 2929.
4189 Also, for "Auto-Submitted", treat anything other than "no" as
4190 non-personal, in accordance with RFC 3834. (Previously it treated
4191 anything starting "auto-" as non-personal.)
4193 TF/01 The control=submission/name=... option had a problem with syntax
4194 errors if the name included a slash character. The /name= option
4195 now slurps the rest of the string, so it can include any characters
4196 but it must come last in the list of options (after /sender_retain
4199 PH/05 Some modifications to the interface to the fake nameserver for the new
4207 TK/01 Added the "success_on_redirect" address verification option. See
4208 NewStuff for rationale and an example.
4210 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
4212 PH/02 Patch to exigrep to allow it to work on syslog lines.
4214 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
4215 fread() to skip over the body file's header line, because in Cygwin the
4216 header line is locked and is inaccessible.
4218 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
4219 co-exist for some time) to make it clear that it is the Exim ID that is
4220 referenced, not the Message-ID: header line.
4222 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
4223 string_format() function, because snprintf() does not exist on all
4226 PH/06 The use of forbid_filter_existstest now also locks out the use of the
4227 ${stat: expansion item.
4229 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
4230 protocol synchronization error", to keep the pedants happy.
4232 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
4233 well as for IRIX systems, when gcc is being used. See the host.c source
4236 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
4238 PH/10 Named domain lists were not working if used in a queue_smtp_domains
4241 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
4242 transport and to the smtp transport in LMTP mode.
4244 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
4246 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
4247 run a filter in a subprocess. This could lead to confusion in subsequent
4248 lookups in the parent process. There should also be a search_tidyup() at
4249 the end of the subprocess.
4251 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
4252 only if the host matched helo_try_verify_hosts, which caused the
4253 verification to occur when the EHLO/HELO command was issued. The ACL just
4254 tested the remembered result. Now, if a previous verification attempt has
4255 not happened, "verify = helo" does it there and then.
4257 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
4259 TK/03 Fix log output including CR from clamd.
4261 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
4262 exist provoked a memory error which could cause a segfault.
4264 PH/15 Installed PCRE 6.2
4266 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
4268 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
4269 of the problem. Specifically, suggested +O2 rather than +O1 for the
4272 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
4274 PH/20 If a delivery was routed to a non-standard port by means of an SRV
4275 record, the port was not correctly logged when the outgoing_port log
4276 selector was set (it logged the transort's default port).
4278 PH/21 Added support for host-specific ports to manualroute, queryprogram,
4279 fallback_hosts, and "hosts" in the smtp transport.
4281 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
4282 host errors such as "Connection refused".
4284 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
4285 authentication with radiusclient 0.4.9:
4287 - Error returned from rc_read_config was caught wrongly
4288 - Username/password not passed on to radius server due to wrong length.
4290 The presumption is that some radiusclient API changes for 4.51/PH/17
4291 were not taken care of correctly. The code is still untested by me (my
4292 Linux distribution still has 0.3.2 of radiusclient), but it was
4293 contributed by a Radius user.
4295 PH/24 When doing a callout, the value of $domain wasn't set correctly when
4296 expanding the "port" option of the smtp transport.
4298 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
4299 while reading a MIME header. Thanks to Tom Hughes for a patch.
4301 PH/24 Include config.h inside local_scan.h so that configuration settings are
4304 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
4305 that in an ACL for RCPT (for example), you can examine exactly what was
4308 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
4309 commands, but it was not correctly comparing the address with the actual
4310 client host address. Thus, it would show the EHLO address in Received:
4311 header lines when this was not necessary.
4313 PH/27 Added the % operator to ${eval:}.
4315 PH/28 Exim tries to create and chdir to its spool directory when it starts;
4316 it should be ignoring failures (because with -C, for example, it has lost
4317 privilege). It wasn't ignoring creation failures other than "already
4320 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
4323 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
4324 because an input line was too long, either on its own, or by virtue of
4325 too many continuations, the temporary file was not being removed, and the
4326 return code was incorrect.
4328 PH/31 Missing "BOOL" in function definition in filtertest.c.
4330 PH/32 Applied Sieve patches from the maintainer.
4332 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
4334 PH/33 Added "verify = not_blind".
4336 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
4337 Local/Makefile (with some defaults set). These are used in built scripts
4338 such as exicyclog, but they have never been used in the exim_install
4339 script (though there are many overriding facilities there). I have
4340 arranged that the exim_install script now takes note of these two
4343 PH/35 Installed configuration files for Dragonfly.
4345 PH/36 When a locally submitted message by a trusted user did not contain a
4346 From: header, and the sender address was obtained from -f or from an SMTP
4347 MAIL command, and the trusted user did not use -F to supply a sender
4348 name, $originator_name was incorrectly used when constructing a From:
4349 header. Furthermore, $originator_name was used for submission mode
4350 messages from external hosts without From: headers in a similar way,
4351 which is clearly wrong.
4353 PH/37 Added control=suppress_local_fixups.
4355 PH/38 When log_selector = +received_sender was set, and the addition of the
4356 sender made the log line's construction buffer exactly full, or one byte
4357 less than full, an overflow happened when the terminating "\n" was
4360 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
4361 when the result of a list match is failure because a DNS lookup failed.
4363 PH/40 RM_COMMAND is now used in the building process.
4365 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
4366 the "build-* directories that it finds.
4368 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
4369 address in a domain literal was a prefix of an interface address.
4371 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
4372 when verifying a sender address, unless rewrite_headers is false.
4374 PH/44 Wrote a long comment about why errors_to addresses are verified as
4375 recipients, not senders.
4377 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
4378 the ratelimit ACL was added.
4380 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
4382 PH/46 Added extra information about PostgreSQL errors to the error string.
4384 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
4385 avoiding the need to install special zones in a real server. This is
4386 backwards compatible; if it can't find the fake resolver, it drops back.
4387 Thus, both old and new test suites can be run.
4389 TF/02 Added util/ratelimit.pl
4391 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
4392 clock is set back in time.
4394 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
4395 Candler <B.Candler@pobox.com>.
4397 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
4398 OK for addresses that are the result of redirections.
4400 PH/48 A number of further additions for the benefit of the new test suite,
4401 including a fake gethostbyname() that interfaces to the fake DNS resolver
4404 TF/06 The fix for widen_domains has also been applied to qualify_single and
4405 search_parents which are the other dnslookup options that can cause
4408 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
4411 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
4413 TF/07 Exim produced the error message "an SRV record indicated no SMTP
4414 service" if it encountered an MX record with an empty target hostname.
4415 The message is now "an MX or SRV record indicated no SMTP service".
4417 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
4418 if the DNS of the sending site is misconfigured. This is quite a
4419 common situation. This change restores the behaviour of treating a
4420 helo verification defer as a failure.
4422 PH/51 If self=fail was set on a router, the bounce message did not include the
4423 actual error message.
4429 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
4431 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
4432 command itself ended in error, the underlying message about the transport
4433 filter timeout was being overwritten with the pipe command error. Now the
4434 underlying error message should be appended to the second error message.
4436 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
4438 PH/02 Reduce the amount of output that "make" produces by default. Full output
4439 can still be requested.
4441 PH/03 The warning log line about a condition test deferring for a "warn" verb
4442 was being output only once per connection, rather than after each
4443 occurrence (because it was using the same function as for successful
4444 "warn" verbs). This seems wrong, so I have changed it.
4446 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
4447 it should not have, which might have caused a crash in the right
4448 circumstances, but probably never did.
4450 PH/04 Installed a modified version of Tony Finch's patch to make submission
4451 mode fix the return path as well as the Sender: header line, and to
4452 add a /name= option so that you can make the user's friendly name appear
4455 TF/03 Added the control = fakedefer ACL modifier.
4457 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
4458 Mark Lowes for thorough testing.
4460 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
4462 TK/03 Merged latest SRS patch from Miles Wilton.
4464 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
4465 with the definition in sysexits.h (which is #included earlier).
4466 Fortunately, Exim does not actually use EX_OK. The code used to try to
4467 preserve the sysexits.h value, by assuming that macro definitions were
4468 scanned for macro replacements. I have been disabused of this notion,
4469 so now the code just undefines EX_OK before #including unistd.h.
4471 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
4472 in the smtp transport. When a block could not be written in a single
4473 write() function, the timeout was being re-applied to each part-write.
4474 This seems wrong - if the receiver was accepting one byte at a time it
4475 would take for ever. The timeout is now adjusted when this happens. It
4476 doesn't have to be particularly precise.
4478 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
4479 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
4481 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
4482 without a domain if the check to <postmaster@domain> fails.
4484 SC/01 Eximstats: added -xls and the ability to specify output files
4485 (patch written by Frank Heydlauf).
4487 SC/02 Eximstats: use FileHandles for outputting results.
4489 SC/03 Eximstats: allow any combination of xls, txt, and html output.
4491 SC/04 Eximstats: fixed display of large numbers with -nvr option
4493 SC/05 Eximstats: fixed merging of reports with empty tables.
4495 SC/06 Eximstats: added the -include_original_destination flag
4497 SC/07 Eximstats: removed tabs and trailing whitespace.
4499 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
4501 TK/06 MBOX spool code: Add real "From " MBOX separator line
4502 so the .eml file is really in mbox format (even though
4503 most programs do not really care). Patch from Alex Miller.
4505 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4506 The latter is generated from $received_to and is only set if the
4507 message has one envelope recipient. SA can use these headers,
4508 obviously out-of-the-box. Patch from Alex Miller.
4510 PH/08 The ${def test on a variable was returning false if the variable's
4511 value was "0", contrary to what the specification has always said!
4512 The result should be true unless the variable is empty.
4514 PH/09 The syntax error of a character other than { following "${if
4515 def:variable_name" (after optional whitespace) was not being diagnosed.
4516 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4517 accidental colon was present, for example, could give incorrect results.
4519 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4520 result is used (not including appendfile, where other changes are about
4523 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4524 This involved changing a lot of size variables from int to off_t. It
4525 should work with maildirs and everything.
4527 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4528 spamd dying while we are connected to it.
4530 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4533 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4534 function for BDB 4.3.
4536 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4538 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4539 the sources for PCRE have been re-arranged and more files are now
4542 PH/15 The code I had for printing potentially long long variables in PH/11
4543 above was not the best (it lost precision). The length of off_t variables
4544 is now inspected at build time, and an appropriate printing format (%ld
4545 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4546 to be "long long int" or "long int". This is needed for the internal
4547 formatting function string_vformat().
4549 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4550 the configuration file to be ":syslog", then the script "guesses" where
4551 the logs files are, rather than using the compiled in default. In our
4552 case the guess is not the same as the compiled default, so the script
4553 suddenly stopped working when I started to use syslog. The patch checks
4554 to see if log_file_path is "". If so, it attempts to read it from exim
4555 with no configuration file to get the compiled in version, before it
4556 falls back to the previous guessing code."
4558 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4559 implementing BATV in an Exim configuration. See NewStuff for the gory
4562 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4563 Makefile that are specific to HP-UX.
4565 PH/18 If the "use_postmaster" option was set for a recipient callout together
4566 with the "random" option, the postmaster address was used as the MAIL
4567 FROM address for the random test, but not for the subsequent recipient
4568 test. It is now used for both.
4570 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4571 patch removes a few documentation additions to RFC 3028, because the
4572 latest draft now contains them. It adds the new en;ascii-case comparator
4573 and a new error check for 8bit text in MIME parts. Comparator and
4574 require names are now matched exactly. I enabled the subaddress
4575 extension, but it is not well tested yet (read: it works for me)."
4577 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4578 rework some of the code of TK/09 above to avoid the hardwired use of
4579 "%lld" and "long long". Replaced the call to snprintf() with a call to
4582 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4583 records point to non-existent hosts", "retry timeout exceeded", and
4584 "retry time not reached for any host after a long failure period".
4586 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4587 experimental DomainKeys support:
4589 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4590 (2) On an error such as an illegally used "control", the wrong name for
4591 the control was given.
4593 These problems did NOT occur unless DomainKeys support was compiled.
4595 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4597 PH/24 Added ${if match_ip condition.
4599 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4600 databases so that it will be absolutely obvious if a crash occurs in the
4601 DB library. This is a regular occurrence (often caused by mis-matched
4604 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4605 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4606 picked up on a user's system that detects such things. There doesn't seem
4607 to be a gcc warning option for this - only an attribute that has to be
4608 put on the function's prototype. It seems that in Fedora Core 4 they have
4609 set this on a number of new functions. No doubt there will be more in due
4612 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4613 specify a transport. However, if an address that was verified by such a
4614 router was the subject of a callout, Exim crashed because it tried to
4615 read the rcpt_include_affixes from the non-existent transport. Now it
4616 just assumes that the setting of that option is false. This bug was
4617 introduced by 4.51/PH/31.
4619 PH/28 Changed -d+all to exclude +memory, because that information is very
4620 rarely of interest, but it makes the output a lot bigger. People tend to
4621 do -d+all out of habit.
4623 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4624 code in os-type was giving problems when libc.so lives in lib64, like on
4627 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4628 aren't the modern standard, and it seems that some systems' include files
4629 don't always have them. Exim was already checking for some of the newer
4630 ones like T_AAAA, and defining it itself. I've added checks for all the
4631 record types that Exim uses.
4633 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4634 not automatically generating a new one, as it is supposed to. This
4635 prevented TLS from working. If the file did exist, but contained invalid
4636 data, a new version was generated, as expected. It was only the case of a
4637 non-existent file that was broken.
4639 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4640 with a change in libdomainkeys > 0.64.
4642 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4643 from DNS. If the selector record carries the flag, it now has
4644 precedence over the domain-wide flag.
4646 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4648 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4649 the use of a port name that isn't defined in /etc/services) occurred, the
4650 message was deferred as in a normal delivery, and thus remained on the
4651 spool, instead of being failed because of the mua_wrapper setting. This
4652 is now fixed, and I tidied up some of the mua_wrapper messages at the
4655 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4656 the messages in a hash of arrays rather than using individual hashes.
4657 This is a bit cleaner and results in dramatic memory savings, albeit
4658 at a slight CPU cost.
4660 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4661 as requested by Marc Sherman.
4663 SC/10 Eximstats: added histograms for user specified patterns as requested
4666 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4668 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4669 fopen() in the content-scanning modules that did not already have it.
4675 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4676 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4678 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4680 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4682 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4683 to test Sieve filters that use "vacation".
4685 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4686 that changes the way the GnuTLS parameters are stored in the cache file.
4687 The new format can be generated externally. For backward compatibility,
4688 if the data in the cache doesn't make sense, Exim assumes it has read an
4689 old-format file, and it generates new data and writes a new file. This
4690 means that you can't go back to an older release without removing the
4693 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4694 work if there are any delivery delays because "one_time" forces the
4695 parent to be marked "delivered", so its unseen clone is never tried
4696 again. For this reason, Exim now forbids the simultaneous setting of
4699 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4700 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4701 entry if you want to know the details. The fix, however, neglected to
4702 consider the case when local delivery batching is involved. The test for
4703 "previously delivered" was not happening when checking to see if an
4704 address could be batched with a previous (undelivered) one; under
4705 certain circumstances this could lead to multiple deliveries to the same
4708 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4709 in its include files, and this causes problems building Exim.
4711 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4712 header_syntax) but Exim was just ignoring anything given after a slash.
4713 In particular, this caused confusion with an attempt to use "verify =
4714 reverse_host_lookup/defer_ok". An error is now given when options are
4715 supplied for verify items that do not have them. (Maybe reverse_host_
4716 lookup should have a defer_ok option, but that's a different point.)
4718 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4719 defined by RFC 821) to 2048, because there were problems with some AUTH
4720 commands, and RFC 1869 says the size should be increased for extended
4721 SMTP commands that take arguments.
4723 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4726 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4727 "unknown" error; now it says that the functionality isn't in the binary.
4729 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4730 an address' error message when a string expansion fails (syntax or
4731 whatever). Otherwise the password may appear in the log. Following change
4732 PH/42 below, there is no longer a chance of it appearing in a bounce
4735 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4737 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4738 of its addresses was ever tried. (Bugzilla bug #2.)
4740 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4741 the result incorrectly in the debug output. (It correctly added a newline
4742 to what was transported.)
4744 TF/01 Added $received_time.
4746 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4747 commented out examples of how to interface to a virus scanner and to
4748 SpamAssassin. Also added commented examples of av_scanner and
4749 spamd_address settings.
4751 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4752 and controls are allowed in which ACLs. There were a couple of minor
4753 errors. Some of the entries in the conditions table (which is a table of
4754 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4755 negation of where the condition IS allowed.
4757 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4759 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4760 header file does not have a version number, so I've had to invent a new
4761 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4762 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4763 radiusclient), but it was contributed by a Radius user.
4765 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4766 files or directories, for OpenSSL.
4768 PH/19 When an Exim process that is running as root has to create an Exim log
4769 file, it does so in a subprocess that runs as exim:exim so as to get the
4770 ownership right at creation (otherwise, other Exim processes might see
4771 the file with the wrong ownership). There was no test for failure of this
4772 fork() call, which would lead to the process getting stuck as it waited
4773 for a non-existent subprocess. Forks do occasionally fail when resources
4774 run out. I reviewed all the other calls to fork(); they all seem to check
4777 PH/20 When checking for unexpected SMTP input at connect time (before writing
4778 the banner), Exim was not dealing correctly with a non-positive return
4779 from the read() function. If the client had disconnected by this time,
4780 the result was a log entry for a synchronization error with an empty
4781 string after "input=" when read() returned zero. If read() returned -1
4782 (an event I could not check), uninitialized data bytes were printed.
4783 There were reports of junk text (parts of files, etc) appearing after
4786 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4788 PH/22 Added support for macro redefinition, and (re)definition in between
4789 driver and ACL definitions.
4791 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4792 forgetting to use the resulting value; it was using the unexpanded value.
4794 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4795 hadn't been configured. The fix is from Juergen Kreileder, who
4796 understands it better than I do:
4798 "Here's what I see happening with three configured cyrus_sasl
4799 authenticators configured (plain, login, cram-md5):
4801 On startup auth_cyrus_sasl_init() gets called for each of these.
4802 This means three calls to sasl_listmech() without a specified mech_list.
4803 => SASL tests which mechs of all available mechs actually work
4804 => three warnings about OTP not working
4805 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4807 With the patch, sasl_listmech() also gets called three times. But now
4808 SASL's mech_list option is set to the server_mech specified in the the
4809 authenticator. Or in other words, the answer from sasl_listmech()
4810 gets limited to just the mech you're testing for (which is different
4812 => the return list contains just 'plain' or 'login', 'cram-md5' or
4813 nothing depending on the value of ob->server_mech.
4815 I've just tested the patch: Authentication still works fine,
4816 unavailable mechs specified in the exim configuration are still
4817 caught, and the auth.log warnings about OTP are gone."
4819 PH/25 When debugging is enabled, the contents of the command line are added
4820 to the debugging output, even when log_selector=+arguments is not
4823 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4824 answer is "GNU", and only if the return is "GNU/something" is the answer
4827 PH/27 $acl_verify_message is now set immediately after the failure of a
4828 verification in an ACL, and so is available in subsequent modifiers. In
4829 particular, the message can be preserved by coding like this:
4831 warn !verify = sender
4832 set acl_m0 = $acl_verify_message
4834 Previously, $acl_verify_message was set only while expanding "message"
4835 and "log_message" when a very denied access.
4837 PH/28 Modified OS/os.c-Linux with
4839 -#ifndef OS_LOAD_AVERAGE
4840 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4842 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4843 nomenclature these days.)
4845 PH/29 Installed patch from the Sieve maintainer that adds the options
4846 sieve_useraddress and sieve_subaddress to the redirect router.
4848 PH/30 In these circumstances:
4849 . Two addresses routed to the same list of hosts;
4850 . First host does not offer TLS;
4851 . First host accepts first address;
4852 . First host gives temporary error to second address;
4853 . Second host offers TLS and a TLS session is established;
4854 . Second host accepts second address.
4855 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4856 and peerdn, if requested) that were in fact used only for the second
4859 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4860 attention to any local part prefix or suffix that was matched by the
4861 router that accepted the address. It now behaves in the same way as it
4862 does for delivery: the affixes are removed from the local part unless
4863 rcpt_include_affixes is set on the transport.
4865 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4866 timeout during the DATA phase of an incoming message.
4868 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4869 applied a patch sanctioned by the Sieve maintainer.
4871 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4872 the uid or gid is negative. A case of a negative gid caused this to be
4873 noticed. The fix allows for either to be negative.
4875 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4876 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4879 PH/36 The change PH/12 above was broken. Fixed it.
4881 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4882 the grounds that routing the same address twice would always produce the
4883 same answer. This might have been true once, but it is certainly no
4884 longer true now. Routing a child address may depend on the previous
4885 routing that produced that child. Some complicated redirection strategies
4886 went wrong when messages had multiple recipients, and made Exim's
4887 behaviour dependent on the order in which the addresses were given.
4889 I have moved the duplicate checking until after the routing is complete.
4890 Exim scans the addresses that are assigned to local and remote
4891 transports, and removes any duplicates. This means that more work will be
4892 done, as duplicates will always all be routed, but duplicates are
4893 presumably rare, so I don't expect this is of any significance.
4895 For deliveries to pipes, files, and autoreplies, the duplicate checking
4896 still happens during the routing process, since they are not going to be
4899 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4900 It corrects a timeout issue with spamd. This is Ian's comment: "The
4901 background is that sometimes spamd either never reads data from a
4902 connection it has accepted, or it never writes response data. The exiscan
4903 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4904 blindly assumes that writes won't block so it may never time out."
4906 PH/39 Allow G after quota size as well as K and M.
4908 PH/40 The value set for $authenticated_id in an authenticator may not contain
4909 binary zeroes or newlines because the value is written to log lines and
4910 to spool files. There was no check on this. Now the value is run through
4911 the string_printing() function so that such characters are converted to
4912 printable escape sequences.
4914 PH/41 $message_linecount is a new variable that contains the total number of
4915 lines in the message. Compare $body_linecount, which is the count for the
4918 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4919 bounce and delay warning messages, except in certain special cases, which
4922 (a) An SMTP error message from a remote host;
4923 (b) A message specified in a :fail: redirection;
4924 (c) A message specified in a "fail" command in a system filter;
4925 (d) A message specified in a FAIL return from the queryprogram router;
4926 (e) A message specified by the cannot_route_message router option.
4928 In these cases only, Exim does include the error details in bounce and
4929 warning messages. There are also a few cases where bland messages such
4930 as "unrouteable address" or "local delivery error" are given.
4932 PH/43 $value is now also set for the "else" part of a ${run expansion.
4934 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4935 being worked on, but at least Exim now implements the latest version to
4938 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4939 process to complete was treated as a delivery failure, a timeout while
4940 writing the message to the pipe was logged, but erroneously treated as a
4941 successful delivery. Such timeouts include transport filter timeouts. For
4942 consistency with the overall process timeout, these timeouts are now
4943 treated as errors, giving rise to delivery failures by default. However,
4944 there is now a new Boolean option for the pipe transport called
4945 timeout_defer, which, if set TRUE, converts the failures into defers for
4946 both kinds of timeout. A transport filter timeout is now identified in
4949 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4950 systems where "make" and "gmake" are different, calling "gmake" at top
4951 level broke things. I've arranged for the value of $(MAKE) to be passed
4952 from the Makefile to this script so that it can call the same version of
4956 A note about Exim versions 4.44 and 4.50
4957 ----------------------------------------
4959 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4960 changes of various kinds. As a consequence, a big documentation update was
4961 needed. This delayed the release for rather longer than seemed good, especially
4962 in the light of a couple of (minor) security issues. Therefore, the changes
4963 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4964 release. So 4.44 and 4.50 are in effect two different branches that both start
4967 I have left the 4.50 change log unchanged; it contains all the changes since
4968 4.43. The change log for 4.44 is below; many of its items are identical to
4969 those for 4.50. This seems to be the most sensible way to preserve the
4970 historical information.
4976 1. Minor wording change to the doc/README.SIEVE file.
4978 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4979 computation of the current number of files was incorrect.
4981 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4982 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4985 4. Give more explanation in the error message when the command for a transport
4986 filter fails to execute.
4988 5. There are several places where Exim runs a non-Exim command in a
4989 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4990 was being done only for the command run by the queryprogram router. It is
4991 now done for all such subprocesses. The other cases are: ${run, transport
4992 filters, and the commands run by the lmtp and pipe transports.
4994 6. Added CONFIGURE_GROUP build-time option.
4996 7. Some older OS have a limit of 256 on the maximum number of file
4997 descriptors. Exim was using setrlimit() to set 1000 as a large value
4998 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4999 systems. I've change it so that if it can't get 1000, it tries for 256.
5001 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
5002 was an oversight, and furthermore, ever since the addition of extra
5003 controls (e.g. 4.43/32), the checks on when to allow different forms of
5004 "control" were broken. There should now be diagnostics for all cases when a
5005 control that does not make sense is encountered.
5007 9. Added the /retain_sender option to "control=submission".
5009 10. $recipients is now available in the predata ACL (oversight).
5011 11. Tidy the search cache before the fork to do a delivery from a message
5012 received from the command line. Otherwise the child will trigger a lookup
5013 failure and thereby defer the delivery if it tries to use (for example) a
5014 cached ldap connection that the parent has called unbind on.
5016 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
5017 of $address_data from the recipient verification was clobbered by the
5018 sender verification.
5020 13. The value of address_data from a sender verification is now available in
5021 $sender_address_data in subsequent conditions in the ACL statement.
5023 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
5025 15. Added a new option "connect=<time>" to callout options, to set a different
5028 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
5029 was its contents. (It was OK if the option was not defined at all.)
5031 17. A "Completed" log line is now written for messages that are removed from
5032 the spool by the -Mrm option.
5034 18. New variables $sender_verify_failure and $recipient_verify_failure contain
5035 information about exactly what failed.
5037 19. Added -dd to debug only the daemon process.
5039 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
5040 handles timeouts, both on the server side and network timeouts. Renamed the
5041 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
5043 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
5044 It is now set to "smtps".
5046 22. $host_address is now set to the target address during the checking of
5047 ignore_target_hosts.
5049 23. When checking ignore_target_hosts for an ipliteral router, no host name was
5050 being passed; this would have caused $sender_host_name to have been used if
5051 matching the list had actually called for a host name (not very likely,
5052 since this list is usually IP addresses). A host name is now passed as
5055 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5056 code that specifies a non-restarting handler (typically sigaction() in
5057 modern systems) in an attempt to fix a rare and obscure crash bug.
5059 25. Narrowed the window for a race in the daemon that could cause it to ignore
5060 SIGCHLD signals. This is not a major problem, because they are used only to
5061 wake it up if nothing else does.
5063 26. A malformed maildirsize file could cause Exim to calculate negative values
5064 for the mailbox size or file count. Odd effects could occur as a result.
5065 The maildirsize information is now recalculated if the size or filecount
5068 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5069 support for a long time. Removed HAVE_SYS_VFS_H.
5071 28. Installed the latest version of exipick from John Jetmore.
5073 29. In an address list, if the pattern was not a regular expression, an empty
5074 subject address (from a bounce message) matched only if the pattern was an
5075 empty string. Non-empty patterns were not even tested. This was the wrong
5076 because it is perfectly reasonable to use an empty address as part of a
5077 database query. An empty address is now tested by patterns that are
5078 lookups. However, all the other forms of pattern expect the subject to
5079 contain a local part and a domain, and therefore, for them, an empty
5080 address still always fails if the pattern is not itself empty.
5082 30. Exim went into a mad DNS loop when attempting to do a callout where the
5083 host was specified on an smtp transport, and looking it up yielded more
5084 than one IP address.
5086 31. Re-factored the code for checking spool and log partition space into a
5087 function that finds that data and another that does the check. The former
5088 is then used to implement four new variables: $spool_space, $log_space,
5089 $spool_inodes, and $log_inodes.
5091 32. The RFC2047 encoding function was originally intended for short strings
5092 such as real names; it was not keeping to the 75-character limit for
5093 encoded words that the RFC imposes. It now respects the limit, and
5094 generates multiple encoded words if necessary. To be on the safe side, I
5095 have increased the buffer size for the ${rfc2047: expansion operator from
5098 33. It is now permitted to omit both strings after an "if" condition; if the
5099 condition is true, the result is "true". As before, when the second string
5100 is omitted, a false condition yields an empty string. This makes it less
5101 cumbersome to write custom ACL and router conditions.
5103 34. Failure to deliver a bounce message always caused it to be frozen, even if
5104 there was an errors_to setting on the router. The errors_to setting is now
5107 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5108 canonical form (fully expanded) before being placed in
5109 $sender_host_address.
5111 36. The table in the code that translates DNS record types into text (T_A to
5112 "A" for instance) was missing entries for NS and CNAME. It is just possible
5113 that this could have caused confusion if both these types were looked up
5114 for the same domain, because the text type is used as part of Exim's
5115 per-process caching. But the chance of anyone hitting this buglet seems
5118 37. The dnsdb lookup has been extended in a number of ways.
5120 (1) There is a new type, "zns", which walks up the domain tree until it
5121 finds some nameserver records. It should be used with care.
5123 (2) There is a new type, "mxh", which is like "mx" except that it returns
5124 just the host names, not the priorities.
5126 (3) It is now possible to give a list of domains (or IP addresses) to be
5127 looked up. The behaviour when one of the lookups defers can be
5128 controlled by a keyword.
5130 (4) It is now possible to specify the separator character for use when
5131 multiple records are returned.
5133 38. The dnslists ACL condition has been extended: it is now possible to supply
5134 a list of IP addresses and/or domains to be looked up in a particular DNS
5137 39. Added log_selector=+queue_time_overall.
5139 40. When running the queue in the test harness, wait just a tad after forking a
5140 delivery process, to get repeatability of debugging output.
5142 41. Include certificate and key file names in error message when GnuTLS fails
5143 to set them up, because the GnuTLS error message doesn't include the name
5144 of the failing file when there is a problem reading it.
5146 42. Allow both -bf and -bF in the same test run.
5148 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
5150 44. The "Exiscan patch" is now merged into the mainline Exim source.
5152 45. Sometimes the final signoff response after QUIT could fail to get
5153 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5154 < 0 before doing a fflush(). This bug looks as though it goes back to the
5155 introduction of TLS in release 3.20, but "sometimes" must have been rare
5156 because the tests only now provoked it.
5158 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
5159 (this can affect the format of dates).
5161 47. exim_tidydb, when checking for the continued existence of a message for
5162 which it has found a message-specific retry record, was not finding
5163 messages that were in split spool directories. Consequently, it was
5164 deleting retry records that should have stayed in existence.
5166 48. Steve fixed some bugs in eximstats.
5168 49. The SPA authentication driver was not abandoning authentication and moving
5169 on to the next authenticator when an expansion was forced to fail,
5170 contradicting the general specification for all authenticators. Instead it
5171 was generating a temporary error. It now behaves as specified.
5173 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
5174 (the order specifies the preference for clients). The order is now AES256,
5175 AES128, 3DES, ARCFOUR128.
5177 51. Small patch to Sieve code - explicitly set From: when generating an
5180 52. Exim crashed if a remote delivery caused a very long error message to be
5181 recorded - for instance if somebody sent an entire SpamAssassin report back
5182 as a large number of 550 error lines. This bug was coincidentally fixed by
5183 increasing the size of one of Exim's internal buffers (big_buffer) that
5184 happened as part of the Exiscan merge. However, to be on the safe side, I
5185 have made the code more robust (and fixed the comments that describe what
5188 53. Now that there can be additional text after "Completed" in log lines (if
5189 the queue_time_overall log selector is set), a one-byte patch to exigrep
5190 was needed to allow it to recognize "Completed" as not the last thing in
5193 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
5194 patch that reportedly fixes this has been added. I am not expert enough to
5195 create a test for it. This is what the patch creator wrote:
5197 "I found a little strange behaviour of ldap code when working with
5198 Windows 2003 AD Domain, where users was placed in more than one
5199 Organization Units. When I tried to give exim partial DN, the exit code
5200 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
5201 But simultaneously result of request was absolutely normal ldap result,
5202 so I produce this patch..."
5204 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
5205 REFERENCE, so I have modified the code to exclude the patch when that macro
5208 55. Some experimental protocols are using DNS PTR records for new purposes. The
5209 keys for these records are domain names, not reversed IP addresses. The
5210 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5211 leaves it alone. Component reversal etc. now happens only for IP addresses.
5214 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
5216 57. Double the size of the debug message buffer (to 2048) so that more of very
5217 long debug lines gets shown.
5219 58. The exicyclog utility now does better if the number of log files to keep
5220 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
5222 59. Two changes related to the smtp_active_hostname option:
5224 (1) $smtp_active_hostname is now available as a variable.
5225 (2) The default for smtp_banner uses $smtp_active_hostname instead
5226 of $primary_hostname.
5228 60. The host_aton() function is supposed to be passed a string that is known
5229 to be a valid IP address. However, in the case of IPv6 addresses, it was
5230 not checking this. This is a hostage to fortune. Exim now panics and dies
5231 if the condition is not met. A case was found where this could be provoked
5232 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5233 components; fortuitously, this particular loophole had already been fixed
5234 by change 4.50/55 above.
5236 If there are any other similar loopholes, the new check in host_aton()
5237 itself should stop them being exploited. The report I received stated that
5238 data on the command line could provoke the exploit when Exim was running as
5239 exim, but did not say which command line option was involved. All I could
5240 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5241 running as the user.
5244 61. There was a buffer overflow vulnerability in the SPA authentication code
5245 (which came originally from the Samba project). I have added a test to the
5246 spa_base64_to_bits() function which I hope fixes it.
5249 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
5250 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
5252 63. The daemon start-up calls getloadavg() while still root for those OS that
5253 need the first call to be done as root, but it missed one case: when
5254 deliver_queue_load_max is set with deliver_drop_privilege. This is
5255 necessary for the benefit of the queue runner, because there is no re-exec
5256 when deliver_drop_privilege is set.
5258 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
5259 This has been fixed.
5261 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5262 was in use, was not putting the data itself into the right store pool;
5263 consequently, it could be overwritten for a subsequent message in the same
5264 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5267 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
5269 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
5270 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
5271 the function were treating the return as a boolean value, which happened to
5272 work because 0=false and not-0=true, but is not correct code.
5274 68. The host_aton() function was not handling scoped IPv6 addresses (those
5275 with, for example, "%eth0" on the end) correctly.
5277 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
5278 negated items (that is, ~something) in unsigned ints. Some compilers
5279 apparently mutter when there is no cast.
5281 70. If an address verification called from an ACL failed, and did not produce a
5282 user-specific message (i.e. there was only a "system" message), nothing was
5283 put in $acl_verify_message. In this situation, it now puts the system
5286 71. Change 4.23/11 added synchronization checking at the start of an SMTP
5287 session; change 4.31/43 added the unwanted input to the log line - except
5288 that it did not do this in the start of session case. It now does.
5290 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
5291 This is wrong and can cause the other end to generate a synchronization
5292 error if it is another Exim or anything else that does the synchronization
5293 check. A QUIT command is no longer sent after a timeout.
5295 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
5296 during host lookups.
5298 74. The defer_ok option of callout verification was not working if it was used
5299 when verifying addresses in header lines, that is, for this case:
5301 verify = header_sender/callout=defer_ok
5303 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
5304 those file descriptors could be used for SMTP connections. If anything
5305 wrote to stderr (the example that came up was "warn" in embedded Perl), it
5306 could be sent to the SMTP client, causing chaos. The daemon now opens
5307 stdin, stdout, and stderr to /dev/null when it puts itself into the
5310 76. Arrange for output from Perl's "warn" command to be written to Exim's main
5311 log by default. The user can override this with suitable Perl magic.
5313 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
5314 the log message when discard triggers, was not working for the DATA ACL or
5315 for the non-SMTP ACL.
5317 78. Error message wording change in sieve.c.
5319 79. If smtp_accept_max_per_host was set, the number of connections could be
5320 restricted to fewer than expected, because the daemon was trying to set up
5321 a new connection before checking whether the processes handling previous
5322 connections had finished. The check for completed processes is now done
5323 earlier. On busy systems, this bug wouldn't be noticed because something
5324 else would have woken the daemon, and it would have reaped the completed
5327 80. If a message was submitted locally by a user whose login name contained one
5328 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
5329 It caused a spool format error. I have fixed the spool reading code. A
5330 related problem was that the "from" clause in the Received: line became
5331 illegal because of the space(s). It is now covered by ${quote_local_part.
5333 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
5336 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
5338 83. Patch from Sieve maintainer to fix unterminated string problem in
5339 "vacation" handling.
5341 84. Some minor changes to the Linux configuration files to help with other
5342 OS variants using glibc.
5344 85. One more patch for Sieve to update vacation handling to latest spec.
5347 ----------------------------------------------------
5348 See the note above about the 4.44 and 4.50 releases.
5349 ----------------------------------------------------
5355 1. Change 4.43/35 introduced a bug that caused file counts to be
5356 incorrectly computed when quota_filecount was set in an appendfile
5359 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
5360 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
5363 3. Give more explanation in the error message when the command for a transport
5364 filter fails to execute.
5366 4. There are several places where Exim runs a non-Exim command in a
5367 subprocess. The SIGUSR1 signal should be disabled for these processes. This
5368 was being done only for the command run by the queryprogram router. It is
5369 now done for all such subprocesses. The other cases are: ${run, transport
5370 filters, and the commands run by the lmtp and pipe transports.
5372 5. Some older OS have a limit of 256 on the maximum number of file
5373 descriptors. Exim was using setrlimit() to set 1000 as a large value
5374 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
5375 systems. I've change it so that if it can't get 1000, it tries for 256.
5377 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
5378 was an oversight, and furthermore, ever since the addition of extra
5379 controls (e.g. 4.43/32), the checks on when to allow different forms of
5380 "control" were broken. There should now be diagnostics for all cases when a
5381 control that does not make sense is encountered.
5383 7. $recipients is now available in the predata ACL (oversight).
5385 8. Tidy the search cache before the fork to do a delivery from a message
5386 received from the command line. Otherwise the child will trigger a lookup
5387 failure and thereby defer the delivery if it tries to use (for example) a
5388 cached ldap connection that the parent has called unbind on.
5390 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
5391 of $address_data from the recipient verification was clobbered by the
5392 sender verification.
5394 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
5395 was its contents. (It was OK if the option was not defined at all.)
5397 11. A "Completed" log line is now written for messages that are removed from
5398 the spool by the -Mrm option.
5400 12. $host_address is now set to the target address during the checking of
5401 ignore_target_hosts.
5403 13. When checking ignore_target_hosts for an ipliteral router, no host name was
5404 being passed; this would have caused $sender_host_name to have been used if
5405 matching the list had actually called for a host name (not very likely,
5406 since this list is usually IP addresses). A host name is now passed as
5409 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5410 code that specifies a non-restarting handler (typically sigaction() in
5411 modern systems) in an attempt to fix a rare and obscure crash bug.
5413 15. Narrowed the window for a race in the daemon that could cause it to ignore
5414 SIGCHLD signals. This is not a major problem, because they are used only to
5415 wake it up if nothing else does.
5417 16. A malformed maildirsize file could cause Exim to calculate negative values
5418 for the mailbox size or file count. Odd effects could occur as a result.
5419 The maildirsize information is now recalculated if the size or filecount
5422 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5423 support for a long time. Removed HAVE_SYS_VFS_H.
5425 18. Updated exipick to current release from John Jetmore.
5427 19. Allow an empty sender to be matched against a lookup in an address list.
5428 Previously the only cases considered were a regular expression, or an
5431 20. Exim went into a mad DNS lookup loop when doing a callout where the
5432 host was specified on the transport, if the DNS lookup yielded more than
5435 21. The RFC2047 encoding function was originally intended for short strings
5436 such as real names; it was not keeping to the 75-character limit for
5437 encoded words that the RFC imposes. It now respects the limit, and
5438 generates multiple encoded words if necessary. To be on the safe side, I
5439 have increased the buffer size for the ${rfc2047: expansion operator from
5442 22. Failure to deliver a bounce message always caused it to be frozen, even if
5443 there was an errors_to setting on the router. The errors_to setting is now
5446 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5447 canonical form (fully expanded) before being placed in
5448 $sender_host_address.
5450 24. Updated eximstats to version 1.33
5452 25. Include certificate and key file names in error message when GnuTLS fails
5453 to set them up, because the GnuTLS error message doesn't include the name
5454 of the failing file when there is a problem reading it.
5456 26. Expand error message when OpenSSL has problems setting up cert/key files.
5459 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
5460 (this can affect the format of dates).
5462 28. exim_tidydb, when checking for the continued existence of a message for
5463 which it has found a message-specific retry record, was not finding
5464 messages that were in split spool directories. Consequently, it was
5465 deleting retry records that should have stayed in existence.
5467 29. eximstats updated to version 1.35
5468 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
5469 1.35 - bugfix such that pie charts by volume are generated correctly
5471 30. The SPA authentication driver was not abandoning authentication and moving
5472 on to the next authenticator when an expansion was forced to fail,
5473 contradicting the general specification for all authenticators. Instead it
5474 was generating a temporary error. It now behaves as specified.
5476 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
5477 (the order specifies the preference for clients). The order is now AES256,
5478 AES128, 3DES, ARCFOUR128.
5480 31. Small patch to Sieve code - explicitly set From: when generating an
5483 32. Exim crashed if a remote delivery caused a very long error message to be
5484 recorded - for instance if somebody sent an entire SpamAssassin report back
5485 as a large number of 550 error lines. This bug was coincidentally fixed by
5486 increasing the size of one of Exim's internal buffers (big_buffer) that
5487 happened as part of the Exiscan merge. However, to be on the safe side, I
5488 have made the code more robust (and fixed the comments that describe what
5491 33. Some experimental protocols are using DNS PTR records for new purposes. The
5492 keys for these records are domain names, not reversed IP addresses. The
5493 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5494 leaves it alone. Component reversal etc. now happens only for IP addresses.
5497 34. The host_aton() function is supposed to be passed a string that is known
5498 to be a valid IP address. However, in the case of IPv6 addresses, it was
5499 not checking this. This is a hostage to fortune. Exim now panics and dies
5500 if the condition is not met. A case was found where this could be provoked
5501 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5502 components; fortuitously, this particular loophole had already been fixed
5503 by change 4.50/55 or 4.44/33 above.
5505 If there are any other similar loopholes, the new check in host_aton()
5506 itself should stop them being exploited. The report I received stated that
5507 data on the command line could provoke the exploit when Exim was running as
5508 exim, but did not say which command line option was involved. All I could
5509 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5510 running as the user.
5513 35. There was a buffer overflow vulnerability in the SPA authentication code
5514 (which came originally from the Samba project). I have added a test to the
5515 spa_base64_to_bits() function which I hope fixes it.
5518 36. The daemon start-up calls getloadavg() while still root for those OS that
5519 need the first call to be done as root, but it missed one case: when
5520 deliver_queue_load_max is set with deliver_drop_privilege. This is
5521 necessary for the benefit of the queue runner, because there is no re-exec
5522 when deliver_drop_privilege is set.
5524 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5525 was in use, was not putting the data itself into the right store pool;
5526 consequently, it could be overwritten for a subsequent message in the same
5527 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5530 38. Sometimes the final signoff response after QUIT could fail to get
5531 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5532 < 0 before doing a fflush(). This bug looks as though it goes back to the
5533 introduction of TLS in release 3.20, but "sometimes" must have been rare
5534 because the tests only now provoked it.
5540 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5541 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5542 Now it returns an integer. A number of calls were still expecting a T/F
5543 return. Fortuitously, in all cases, the tests worked in OK situations,
5544 which is the norm. However, things would have gone wrong on any write
5545 failures on the smtp file descriptor. This function is used when sending
5546 messages over SMTP and also when doing verify callouts.
5548 2. When Exim is called to do synchronous delivery of a locally submitted
5549 message (the -odf or -odi options), it no longer closes stderr before doing
5552 3. Implemented the mua_wrapper option.
5554 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5556 5. Implemented the functions header_remove(), header_testname(),
5557 header_add_at_position(), and receive_remove_recipient(), and exported them
5560 6. If an ACL "warn" statement specified the addition of headers, Exim already
5561 inserted X-ACL-Warn: at the start if there was no header name. However, it
5562 was not making this test for the second and subsequent header lines if
5563 there were newlines in the string. This meant that an invalid header could
5564 be inserted if Exim was badly configured.
5566 7. Allow an ACL "warn" statement to add header lines at the start or after all
5567 the Received: headers, as well as at the end.
5569 8. Added the rcpt_4xx retry error code.
5571 9. Added postmaster_mailfrom=xxx to callout verification option.
5573 10. Added mailfrom=xxxx to the callout verification option, for verify=
5576 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5577 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5579 12. Inserted some casts to stop certain compilers warning when using pointer
5580 differences as field lengths or precisions in printf-type calls (mostly
5581 affecting debugging statements).
5583 13. Added optional readline() support for -be (dynamically loaded).
5585 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5586 same clock tick as a message's arrival, so that its received time was the
5587 same as the "first fail" time on the retry record, and that message
5588 remained on the queue past the ultimate address timeout, every queue runner
5589 would try a delivery (because it was past the ultimate address timeout) but
5590 after another failure, the ultimate address timeout, which should have then
5591 bounced the address, did not kick in. This was a "< instead of <=" error;
5592 in most cases the first failure would have been in the next clock tick
5593 after the received time, and all would be well.
5595 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5596 being recognized when the domain list was tested by the match_domain
5597 condition in an expansion string.
5599 16. Added the ${str2b64: operator.
5601 17. Exim was always calling setrlimit() to set a large limit for the number of
5602 processes, without checking whether the existing limit was already
5603 adequate. (It did check for the limit on file descriptors.) Furthermore,
5604 errors from getrlimit() and setrlimit() were being ignored. Now they are
5605 logged to the main and panic logs, but Exim does carry on, to try to do its
5606 job under whatever limits there are.
5608 18. Imported PCRE 5.0.
5610 19. Trivial typo in log message " temporarily refused connection" (the leading
5613 20. If the log selector return_path_on_delivery was set and an address was
5614 redirected to /dev/null, the delivery process crashed because it assumed
5615 that a return path would always be set for a "successful" delivery. In this
5616 case, the whole delivery is bypassed as an optimization, and therefore no
5619 21. Internal re-arrangement: the function for sending a challenge and reading
5620 a response while authentication was assuming a zero-terminated challenge
5621 string. It's now changed to take a pointer and a length, to allow for
5622 binary data in such strings.
5624 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5626 23. Exim was not respecting finduser_retries when seeking the login of the
5627 uid under which it was called; it was always trying 10 times. (The default
5628 setting of finduser_retries is zero.) Also, it was sleeping after the final
5629 failure, which is pointless.
5631 24. Implemented tls_on_connect_ports.
5633 25. Implemented acl_smtp_predata.
5635 26. If the domain in control=submission is set empty, Exim assumes that the
5636 authenticated id is a complete email address when it generates From: or
5637 Sender: header lines.
5639 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5640 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5641 chown and chgrp in /bin and hostname in /usr/bin.
5643 28. Exim was keeping the "process log" file open after each use, just as it
5644 does for the main log. This opens the possibility of it remaining open for
5645 long periods when the USR1 signal hits a daemon. Occasional processlog
5646 errors were reported, that could have been caused by this. Anyway, it seems
5647 much more sensible not to leave this file open at all, so that is what now
5650 29. The long-running daemon process does not normally write to the log once it
5651 has entered its main loop, and it closes the log before doing so. This is
5652 so that log files can straightforwardly be renamed and moved. However,
5653 there are a couple of unusual error situations where the daemon does write
5654 log entries, and I had neglected to close the log afterwards.
5656 30. The text of an SMTP error response that was received during a remote
5657 delivery was being truncated at 512 bytes. This is too short for some of
5658 the long messages that one sometimes sees. I've increased the limit to
5661 31. It is now possible to make retry rules that apply only when a message has a
5662 specific sender, in particular, an empty sender.
5664 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5665 it possible to be selective about when SMTP synchronization is enforced.
5667 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5669 32. Implemented hosts_connection_nolog.
5671 33. Added an ACL for QUIT.
5673 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5676 35. Added mailbox_size and mailbox_filecount to appendfile.
5678 36. Added control = no_multiline_responses to ACLs.
5680 37. There was a bug in the logic of the code that waits for the clock to tick
5681 in the case where the clock went backwards by a substantial amount such
5682 that the microsecond fraction of "now" was more than the microsecond
5683 fraction of "then" (but the whole seconds number was less).
5685 38. Added support for the libradius Radius client library this is found on
5686 FreeBSD (previously only the radiusclient library was supported).
5692 1. When certain lookups returned multiple values in the form name=value, the
5693 quoting of the values was not always being done properly. Specifically:
5694 (a) If the value started with a double quote, but contained no whitespace,
5696 (b) If the value contained whitespace other than a space character (i.e.
5697 tabs or newlines or carriage returns) it was not quoted.
5698 This fix has been applied to the mysql and pgsql lookups by writing a
5699 separate quoting function and calling it from the lookup code. The fix
5700 should probably also be applied to nisplus, ibase and oracle lookups, but
5701 since I cannot test any of those, I have not disturbed their existing code.
5703 2. A hit in the callout cache for a specific address caused a log line with no
5704 reason for rejecting RCPT. Now it says "Previous (cached) callout
5705 verification failure".
5707 3. There was an off-by-one bug in the queryprogram router. An over-long
5708 return line was truncated at 256 instead of 255 characters, thereby
5709 overflowing its buffer with the terminating zero. As well as fixing this, I
5710 have increased the buffer size to 1024 (and made a note to document this).
5712 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5713 when Exim is waiting for an SMTP response from a remote server, Exim
5714 restarts its select() call on the socket, thereby resetting its timeout.
5715 This is not a problem when such interrupts are rare. Somebody set up a cron
5716 job to run exiwhat every 2 minutes, which is less than the normal select()
5717 timeout (5 or 10 minutes). This meant that the select() timeout never
5718 kicked in because it was always reset. I have fixed this by comparing the
5719 time when an interrupt arrives with the time at the start of the first call
5720 to select(). If more time than the timeout has elapsed, the interrupt is
5721 treated as a timeout.
5723 5. Some internal re-factoring in preparation for the addition of Sieve
5724 extensions (by MH). In particular, the "personal" test is moved to a
5725 separate function, and given an option for scanning Cc: and Bcc: (which is
5726 not set for Exim filters).
5728 6. When Exim created an email address using the login of the caller as the
5729 local part (e.g. when creating a From: or Sender: header line), it was not
5730 quoting the local part when it contained special characters such as @.
5732 7. Installed new OpenBSD configuration files.
5734 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5735 try to make them clearer.
5737 9. Callout options, other than the timeout value, were being ignored when
5738 verifying sender addresses in header lines. For example, when using
5740 verify = header_sender/callout=no_cache
5742 the cache was (incorrectly) being used.
5744 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5745 only the Cygwin environment.
5747 11. When return_path_on_delivery was set as a log selector, if different remote
5748 addresses in the same message used different return paths and parallel
5749 remote delivery occurred, the wrong values would sometimes be logged.
5750 (Whenever a remote delivery process finished, the return path value from
5751 the most recently started remote delivery process was logged.)
5753 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5754 lines when AUTH and/or TLS are in use. This is the "received protocol"
5755 field. Exim used to use "asmtp" for authenticated SMTP, without any
5756 indication (in the protocol name) for TLS use. Now it follows the RFC and
5757 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5758 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5759 names appear in log lines as well as in Received: header lines.
5761 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5762 extensions, and comparison tests, and to fix some bugs.
5764 14. Changes to the "personal" filter test:
5766 (1) The test was buggy in that it was just doing the equivalent of
5767 "contains" tests on header lines. For example, if a user's address was
5768 anne@some.where, the "personal" test would incorrectly be true for
5770 To: susanne@some.where
5772 This test is now done by extracting each address from the header in turn,
5773 and checking the entire address. Other tests that are part of "personal"
5774 are now done using regular expressions (for example, to check local parts
5775 of addresses in From: header lines).
5777 (2) The list of non-personal local parts in From: addresses has been
5778 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5779 taken from the Sieve specification recommendations.
5781 (3) If the message contains any header line starting with "List-" it is
5782 treated as non-personal.
5784 (4) The test for "circular" in the Subject: header line has been removed
5785 because it now seems ill-conceived.
5787 15. Minor typos in src/EDITME comments corrected.
5789 16. Installed latest exipick from John Jetmore.
5791 17. If headers_add on a router specified a text string that was too long for
5792 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5793 of string_sprintf() is now avoided.
5795 18. $message_body_size was not set (it was always zero) when running the DATA
5796 ACL and the local_scan() function.
5798 19. For the "mail" command in an Exim filter, no default was being set for
5799 the once_repeat time, causing a random time value to be used if "once" was
5800 specified. (If the value happened to be <= 0, no repeat happened.) The
5801 default is now 0s, meaning "never repeat". The "vacation" command was OK
5802 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5803 (I found it when inspecting the code).
5805 20. There is now an overall timeout for performing a callout verification. It
5806 defaults to 4 times the callout timeout, which applies to individual SMTP
5807 commands during the callout. The overall timeout applies when there is more
5808 than one host that can be tried. The timeout is checked before trying the
5809 next host. This prevents very long delays if there are a large number of
5810 hosts and all are timing out (e.g. when the network connections are timing
5811 out). The value of the overall timeout can be changed by specifying an
5812 additional sub-option for "callout", called "maxwait". For example:
5814 verify = sender/callout=5s,maxwait=20s
5816 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5817 to the end before writing, but this should make it even safer).
5819 22. Exim was forgetting that it had advertised PIPELINING for the second and
5820 subsequent messages on an SMTP connection. It was also not resetting its
5821 memory on STARTTLS and an internal HELO.
5823 23. When Exim logs an SMTP synchronization error within a session, it now
5824 records whether PIPELINING has been advertised or not.
5826 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5827 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5828 rather than long int.
5830 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5832 26. Added the never_mail option to autoreply.
5838 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5839 crash if the getsockname() call failed; this can happen if a connection is
5840 closed very soon after it is established. The problem was simply in the
5841 order in which certain operations were done, causing Exim to try to write
5842 to the SMTP stream before it had set up the file descriptor. The bug has
5843 been fixed by making things happen in the correct order.
5849 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5850 before the connection was closed, thus losing the rejection response.
5852 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5853 some early Solaris releases, but causes trouble in current releases where
5854 socklen_t is defined.
5856 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5859 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5862 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5863 that releasing the top of it at the end releases what was used for sub-
5864 expansions (unless the block got too big). However, discard this block if
5865 the first thing is a variable or header, so that we can use its block when
5866 it is dynamic (useful for very large $message_headers, for example).
5868 6. Lookups now cache *every* query, not just the most recent. A new, separate
5869 store pool is used for this. It can be recovered when all lookup caches are
5870 flushed. Lookups now release memory at the end of their result strings.
5871 This has involved some general refactoring of the lookup sources.
5873 7. Some code has been added to the store_xxx() functions to reduce the amount
5874 of flapping under certain conditions.
5876 8. log_incoming_interface used to affect only the <= reception log lines. Now
5877 it causes the local interface and port to be added to several more SMTP log
5878 lines, for example "SMTP connection from", and rejection lines.
5880 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5882 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5884 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5885 could be overwritten at the end of the current message (or the start of a
5886 new message if it was set in a HELO ACL). The value is now preserved for
5887 the duration of the SMTP connection.
5889 12. If a transport had a headers_rewrite setting, and a matching header line
5890 contained an unqualified address, that address was qualified, even if it
5891 did not match any rewriting rules. The underlying bug was that the values
5892 of the flags that permit the existence of unqualified sender and recipient
5893 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5894 non-local messages, and by -bnq for local messages) were not being
5895 preserved with the message after it was received.
5897 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5898 "next input=" as part of the text comprising the host identity instead of
5899 the correct text. The code was using the same buffer for two different
5900 strings. However, depending on which order the printing function evaluated
5901 its arguments, the bug did not always show up. Under Linux, for example, my
5902 test suite worked just fine.
5904 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5905 doesn't work with some older versions of Perl. It has been changed to "my",
5906 which in any case is probably the better facility to use.
5908 15. A really picky compiler found some instances of statements for creating
5909 error messages that either had too many or two few arguments for the format
5912 16. The size of the buffer for calls to the DNS resolver has been increased
5913 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5914 for addresses that have a lot of PTR records. This alleviates a problem; it
5915 does not fully solve it.
5917 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5918 buffer now truncates the list and logs the incident, which is the same
5919 action as happens when Exim is looking up a host name and its aliases.
5920 Previously in this situation something unpredictable would happen;
5921 sometimes it was "internal error: store_reset failed".
5923 18. If a server dropped the connection unexpectedly when an Exim client was
5924 using GnuTLS and trying to read a response, the client delivery process
5925 crashed while trying to generate an error log message.
5927 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5928 string, for example:
5930 warn message = H1: something\nH2: something
5932 the text was added as a single header line from Exim's point of view
5933 though it ended up OK in the delivered message. However, searching for the
5934 second and subsequent header lines using $h_h2: did not work. This has been
5935 fixed. Similarly, if a system filter added multiple headers in this way,
5936 the routers could not see them.
5938 20. Expanded the error message when iplsearch is called with an invalid key to
5939 suggest using net-iplsearch in a host list.
5941 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5942 ACLs are no longer actually imposed (and a message to that effect is
5945 22. If a "gecos" field in a passwd entry contained escaped characters, in
5946 particular, if it contained a \" sequence, Exim got it wrong when building
5947 a From: or a Sender: header from that name. A second bug also caused
5948 incorrect handling when an unquoted " was present following a character
5949 that needed quoting.
5951 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5952 was not being matched caselessly.
5954 24. Arranged for all hyphens in the exim.8 source to be escaped with
5957 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5958 properly. Recipient callout cache records were still being keyed to include
5959 the sender, even when use_sender was set false. This led to far more
5960 callouts that were necessary. The sender is no longer included in the key
5961 when use_sender is false.
5963 26. Added "control = submission" modifier to ACLs.
5965 27. Added the ${base62d: operator to decode base 62 numbers.
5967 28. dnsdb lookups can now access SRV records.
5969 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5970 the configuration file.
5972 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5973 (-v) mode. This makes the output for a verbose queue run more intelligible.
5975 31. Added a use_postmaster feature to recipient callouts.
5977 32. Added the $body_zerocount variable, containing the number of binary zero
5978 bytes in the message body.
5980 33. The time of last modification of the "new" subdirectory is now used as the
5981 "mailbox time last read" when there is a quota error for a maildir
5984 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5986 35. Added +ignore_unknown as a special item in host lists.
5988 36. Code for decoding IPv6 addresses in host lists is now included, even if
5989 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5990 address was recognized as an IP address, but was then not correctly decoded
5991 into binary, causing unexpected and incorrect effects when compared with
5998 1. Very minor rewording of debugging text in manualroute to say "list of
5999 hosts" instead of "hostlist".
6001 2. If verify=header_syntax was set, and a header line with an unqualified
6002 address (no domain) and a large number of spaces between the end of the
6003 name and the colon was received, the reception process suffered a buffer
6004 overflow, and (when I tested it) crashed. This was caused by some obsolete
6005 code that should have been removed. The fix is to remove it!
6007 3. When running in the test harness, delay a bit after writing a bounce
6008 message to get a bit more predictability in the log output.
6010 4. Added a call to search_tidyup() just before forking a reception process. In
6011 theory, someone could use a lookup in the expansion of smtp_accept_max_
6012 per_host which, without the tidyup, could leave open a database connection.
6014 5. Added the variables $recipient_data and $sender_data which get set from a
6015 lookup success in an ACL "recipients" or "senders" condition, or a router
6016 "senders" option, similar to $domain_data and $local_part_data.
6018 6. Moved the writing of debug_print from before to after the "senders" test
6021 7. Change 4.31/66 (moving the time when the Received: is generated) caused
6022 problems for message scanning, either using a data ACL, or using
6023 local_scan() because the Received: header was not generated till after they
6024 were called (in order to set the time as the time of reception completion).
6025 I have revised the way this works. The header is now generated after the
6026 body is received, but before the ACL or local_scan() are called. After they
6027 are run, the timestamp in the header is updated.
6033 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
6034 before starting a queue runner without re-exec. This happened only when
6035 deliver_drop_privilege was set or when the Exim user was set to root. The
6036 effect of the bug was that timeouts during subsequent deliveries caused
6037 crashes instead of being properly handled. The handler is now left at its
6038 default (and expected) setting.
6040 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
6041 message, again when deliver_drop_privilege is set or Exim is run as root.
6042 The bug described in (1) was not present in this case, but the tidying up
6043 of the other signals was missing. I have made the two cases consistent.
6045 3. The ignore_target_hosts setting on a manualroute router was being ignored
6046 for hosts that were looked up using the /MX notation.
6048 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
6051 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
6052 operated on the sender address. After changing the $sender_address to <>
6053 for the sender address verify, Exim was re-instated it as the original
6054 (before rewriting) address, but remembering that it had rewritten it, so it
6055 wasn't rewriting it again. This bug also had the effect of breaking the
6056 sender address verification caching when the sender address was rewritten.
6058 6. The ignore_target_hosts option was being ignored by the ipliteral router.
6059 This has been changed so that if the ip literal address matches
6060 ignore_target_hosts, the router declines.
6062 7. Added expansion conditions match_domain, match_address, and match_local_
6063 part (NOT match_host).
6065 8. The placeholder for the Received: header didn't have a length field set.
6067 9. Added code to Exim itself and to exim_lock to test for a specific race
6068 condition that could lead to file corruption when using MBX delivery. The
6069 issue is with the lockfile that is created in /tmp. If this file is removed
6070 after a process has opened it but before that process has acquired a lock,
6071 there is the potential for a second process to recreate the file and also
6072 acquire a lock. This could lead to two Exim processes writing to the file
6073 at the same time. The added code performs the same test as UW imapd; it
6074 checks after acquiring the lock that its file descriptor still refers to
6075 the same named file.
6077 10. The buffer for building added header lines was of fixed size, 8192 bytes.
6078 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
6081 11. Added the smtp_active_hostname option. If used, this will typically be made
6082 to depend on the incoming interface address. Because $interface_address is
6083 not set up until the daemon has forked a reception process, error responses
6084 that can happen earlier (such as "too many connections") no longer contain
6087 12. If an expansion in a condition on a "warn" statement fails because a lookup
6088 defers, the "warn" statement is abandoned, and the next ACL statement is
6089 processed. Previously this caused the whole ACL to be aborted.
6091 13. Added the iplsearch lookup type.
6093 14. Added ident_timeout as a log selector.
6095 15. Added tls_certificate_verified as a log selector.
6097 16. Added a global option tls_require_ciphers (compare the smtp transport
6098 option of the same name). This controls incoming TLS connections.
6100 17. I finally figured out how to make tls_require_ciphers do a similar thing
6101 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
6102 before starting the TLS session.
6104 18. Tabs are now shown as \t in -bP output.
6106 19. If the log selector return_path_on_delivery was set, Exim crashed when
6107 bouncing a message because it had too many Received: header lines.
6109 20. If two routers both had headers_remove settings, and the first one included
6110 a superfluous trailing colon, the final name in the first list and the
6111 first name in the second list were incorrectly joined into one item (with a
6112 colon in the middle).
6118 1. Added -C and -D options to the exinext utility, mainly to make it easier
6119 to include in the automated testing, but these could be helpful when
6120 multiple configurations are in use.
6122 2. The exinext utility was not formatting the output nicely when there was
6123 an alternate port involved in the retry record key, nor when there was a
6124 message id as well (for retries that were specific to a specific message
6125 and a specific host). It was also confused by IPv6 addresses, because of
6126 the additional colons they contain. I have fixed the IPv4 problem, and
6127 patched it up to do a reasonable job for IPv6.
6129 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
6130 delivery, the log line now contains "pipelined" if PIPELINING was used.
6132 4. An SMTP transport process used to panic and die if the bind() call to set
6133 an explicit outgoing interface failed. This has been changed; it is now
6134 treated in the same way as a connect() failure.
6136 5. A reference to $sender_host_name in the part of a conditional expansion
6137 that was being skipped was still causing a DNS lookup. This no longer
6140 6. The def: expansion condition was not recognizing references to header lines
6141 that used bh_ and bheader_.
6143 7. Added the _cache feature to named lists.
6145 8. The code for checking quota_filecount in the appendfile transport was
6146 allowing one more file than it should have been.
6148 9. For compatibility with Sendmail, the command line option
6156 and sets the incoming protocol and host name (for trusted callers). The
6157 host name and its colon can be omitted when only the protocol is to be set.
6158 Note the Exim already has two private options, -pd and -ps, that refer to
6159 embedded Perl. It is therefore impossible to set a protocol value of "d" or
6160 "s", but I don't think that's a major issue.
6162 10. A number of refactoring changes to the code, none of which should affect
6165 (a) The number of logging options was getting close to filling up the
6166 32-bit word that was used as a bit map. I have split them into two classes:
6167 those that are passed in the argument to log_write(), and those that are
6168 only ever tested independently outside of that function. These are now in
6169 separate 32-bit words, so there is plenty of room for expansion again.
6170 There is no change in the user interface or the logging behaviour.
6172 (b) When building, for example, log lines, the code previously used a
6173 macro that called string_cat() twice, in order to add two strings. This is
6174 not really sufficiently general. Furthermore, there was one instance where
6175 it was actually wrong because one of the argument was used twice, and in
6176 one call a function was used. (As it happened, calling the function twice
6177 did not affect the overall behaviour.) The macro has been replaced by a
6178 function that can join an arbitrary number of extra strings onto a growing
6181 (c) The code for expansion conditions now uses a table and a binary chop
6182 instead of a serial search (which was left over from when there were very
6183 few conditions). Also, it now recognizes conditions like "pam" even when
6184 the relevant support is not compiled in: a suitably worded error message is
6185 given if an attempt is made to use such a condition.
6187 11. Added ${time_interval:xxxxx}.
6189 12. A bug was causing one of the ddress fields not to be passed back correctly
6190 from remote delivery subprocesses. The field in question was not being
6191 subsequently used, so this caused to problems in practice.
6193 13. Added new log selectors queue_time and deliver_time.
6195 14. Might have fixed a bug in maildirsizefile handling that threw up
6196 "unexpected character" debug warnings, and recalculated the data
6197 unnecessarily. In any case, I expanded the warning message to give more
6200 15. Added the message "Restricted characters in address" to the statements in
6201 the default ACL that block characters like @ and % in local parts.
6203 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
6204 Three changes have been made:
6206 (a) There was a serious bug; a negative response to MAIL caused the whole
6207 recipient domain to be cached as invalid, thereby blocking all messages
6208 to all local parts at the same domain, from all senders. This bug has
6209 been fixed. The domain is no longer cached after a negative response to
6210 MAIL if the sender used is not empty.
6212 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
6215 (c) A new callout option, "use_sender" has been added for people who want
6216 the modified behaviour.
6222 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
6225 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
6226 indeed breaks things for older releases.
6228 3. Added additional logging to the case where there is a problem reading data
6229 from a filter that is running in a subprocess using a pipe, in order to
6230 try to track down a specific problem.
6232 4. Testing facility fudge: when running in the test harness and attempting
6233 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
6234 getting "No route to host". Convert this to a timeout.
6236 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
6239 6. Some OS don't have socklen_t but use size_t instead. This affects the
6240 fifth argument of getsockopt() amongst other things. This is now
6241 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
6242 can be set for individual OS. I have set it for SunOS5, OSF1, and
6243 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
6244 some earlier ones do not.
6246 7. Change 4.30/15 was not doing the test caselessly.
6248 8. The standard form for an IPv6 address literal was being rejected by address
6249 parsing in, for example, MAIL and RCPT commands. An example of this kind of
6250 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6251 this, as well as the form without the "IPv6" on the front (but only when
6252 address literals are enabled, of course).
6254 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
6256 10. Exim crashed if a message with an empty sender address specified by -f
6257 encountered a router with an errors_to setting. This could be provoked only
6258 by a command such as
6262 where an empty string was supplied; "<>" did not hit this bug.
6264 11. Installed PCRE release 4.5.
6266 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
6267 remained set. It is now erased.
6269 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
6270 times from message ids (which are base 36 rather than the normal 62).
6272 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
6273 were being counted as actual protocol errors, and logged if the log
6274 selector +smtp_protocol_error was set. One cannot be perfect in this test,
6275 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
6276 and DATA following a set of rejected RCPTs do not count as protocol errors.
6277 In other words, Exim assumes they were pipelined, though this may not
6278 actually be the case. Of course, in all cases the client gets an
6279 appropriate error code.
6281 15. If a lookup fails in an ACL condition, a message about the failure may
6282 be available; it is used if testing the ACL cannot continue, because most
6283 such messages specify what the cause of the deferral is. However, some
6284 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
6285 that caused an old message to be retained and used if a later statement
6286 caused a defer, replacing the real cause of the deferral.
6288 16. If an IP address had so many PTR records that the DNS lookup buffer
6289 was not large enough to hold them, Exim could crash while trying to process
6290 the truncated data. It now detects and logs this case.
6292 17. Further to 4.21/58, another change has been made: if (and only if) the
6293 first line of a message (the first header line) ends with CRLF, a bare LF
6294 in a subsequent header line has a space inserted after it, so as not to
6295 terminate the header.
6297 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
6298 unnecessarily, used atoi() instead of strtol(), and didn't check the
6299 termination when getting file sizes from file names by regex.
6301 19. Completely re-implemented the support for maildirsize files, in the light
6302 of a number of problems with the previous contributed implementation
6303 (4.30/29). In particular:
6305 . If the quota is zero, the maildirsize file is maintained, but no quota is
6308 . If the maildir directory does not exist, it is created before any attempt
6309 to write a maildirsize file.
6311 . The quota value in the file is just a cache; if the quota is changed in
6312 the transport, the new value overrides.
6314 . A regular expression is available for excluding directories from the
6317 20. The autoreply transport checks the characters in options that define the
6318 message's headers; it allows continued headers, but it was checking with
6319 isspace() after an embedded newline instead of explicitly looking for a
6322 21. If all the "regular" hosts to which an address was routed had passed their
6323 expiry times, and had not reached their retry times, the address was
6324 bounced, even if fallback hosts were defined. Now Exim should go on to try
6327 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
6328 equivalent code in the SMTP transport. Some hosts send humungous responses
6329 to HELO/EHLO, more than 1024 it seems.
6331 23. Refactoring: code in filter.c used (void *) for "any old type" but this
6332 gives compiler warnings in some environments. I've now done it "properly",
6335 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
6336 (because of problems with the built-in one) was declared to return uschar *
6337 instead of char *, causing compiler failure.
6339 25. Fixed a file descriptor leak when processing alias/forward files.
6341 26. Fixed a minor format string issue in dbfn.c.
6343 27. Typo in exim.c: ("dmbnz" for "dbmnz").
6345 28. If a filter file refered to $h_xxx or $message_headers, and the headers
6346 contained RFC 2047 "words", Exim's memory could, under certain conditions,
6349 29. When a sender address is verified, it is cached, to save repeating the test
6350 when there is more than one recipient in a message. However, when the
6351 verification involves a callout, it is possible for different callout
6352 options to be set for different recipients. It is too complicated to keep
6353 track of this in the cache, so now Exim always runs a verification when a
6354 callout is required, relying on the callout cache for the optimization.
6355 The overhead is duplication of the address routing, but this should not be
6358 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
6359 to be verified with callout=postmaster, and the main callout worked but the
6360 postmaster check failed, the verification correctly failed. However, if a
6361 subsequent RCPT command asked for sender verification *without* the
6362 postmaster check, incorrect caching caused this verification also to fail,
6365 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
6366 it was not caching the DNS options (qualify_single, search_parents) that
6367 were used when the lookup failed. A subsequent lookup with different
6368 options therefore always gave the same answer, though there were cases
6369 where it should not have. (Example: a "domains = !$mx_any" option on a
6370 dnslookup router: the "domains" option is always processed without any
6371 widening, but the router might have qualify_single set.) Now Exim uses the
6372 cached value only when the same options are set.
6374 32. Added John Jetmore's "exipick" utility to the distribution.
6376 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
6377 than a timeout (e.g. a certificate is required, and is not provided), an
6378 Exim server now closes the connection immediately. Previously it waited for
6379 the client to close - but if the client is SSL, it seems that they each
6380 wait for each other, leading to a delay before one of them times out.
6382 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
6383 maintained 0.8.x compatibility because I don't think many are using it, and
6384 it is clearly obsolete.
6386 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
6389 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
6390 client certificate was expired. A simple patch fixes this, though I don't
6391 understand the full logic of why the verify callback is called multiple
6394 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
6395 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
6396 which causes problems with some clients (such as the Certicom SSL Plus
6397 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
6398 disables the coutermeasure allowing Eudora to connect."
6400 38. Exim was not checking that a write() to a log file succeeded. This could
6401 lead to Bad Things if a log got too big, in particular if it hit a file
6402 size limit. Exim now panics and dies if it cannot write to a log file, just
6403 as it does if it cannot open a log file.
6405 39. Modified OS/Makefile-Linux so that it now contains
6407 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
6409 The two -D definitions ensure that Exim is compiled with large file
6410 support, which makes it possible to handle log files that are bigger than
6413 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
6414 instance) a domain was checked against a named list that involved a lookup,
6415 causing $domain_data to be set, then another domain was checked against the
6416 same list, then the first domain was re-checked, the value of $domain_data
6417 after the final check could be wrong. In particular, if the second check
6418 failed, it could be set empty. This bug probably also applied to
6421 41. The strip_trailing_dot option was not being applied to the address given
6422 with the -f command-line option.
6424 42. The code for reading a message's header from the spool was incrementing
6425 $received_count, but never initializing it. This meant that the value was
6426 incorrect (doubled) while delivering a message in the same process in which
6427 it was received. In the most common configuration of Exim, this never
6428 happens - a fresh exec is done - but it can happen when
6429 deliver_drop_privilege is set.
6431 43. When Exim logs an SMTP synchronization error - client data sent too soon -
6432 it now includes up to 150 characters of the unexpected data in the log
6435 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
6436 and building data strings. The size of both of these buffers was 10 000
6437 bytes - far larger than anybody would *ever* want, thought I. Needless to
6438 say, somebody hit the limit. I have increased the maximum line length to
6439 20 000 and the maximum data length of concatenated lines to 100 000. I have
6440 also fixed two bugs, because there was no checking on these buffers. Tsk,
6441 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
6442 buffer is too small.
6444 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
6445 lsearch lookups. Now it does.
6447 46. When parsing a route_list item in a manualroute router, a fixed-length
6448 buffer was used for the list of hosts. I made this 1024 bytes long,
6449 thinking that nobody would ever have a list of hosts that long. Wrong.
6450 Somebody had a whole pile of complicated expansion conditions, and the
6451 string was silently truncated, leading to an expansion error. It turns out
6452 that it is easier to change to an unlimited length (owing to other changes
6453 that have happened since this code was originally written) than to build
6454 structure for giving a limitation error. The length of the item that
6455 expands into the list of hosts is now unlimited.
6457 47. The lsearch lookup could not handle data where the length of text line was
6458 more than 4095 characters. Such lines were truncated, leading to shortened
6459 data being returned. It should now handle lines of any length.
6461 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
6462 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
6465 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
6466 replacing strings such as BIN_DIRECTORY in a source file: the replacement
6467 no longer happens in comment lines. A list of replacements is now placed
6468 at the head of all of the source files, except those whose only change is
6469 to replace PERL_COMMAND in the very first #! line.
6471 50. Replaced the slow insertion sort in queue.c, for sorting the list of
6472 messages on the queue, with a bottom-up merge sort, using code contributed
6473 by Michael Haardt. This should make operations like -bp somewhat faster on
6474 large queues. It won't affect queue runners, except when queue_run_in_order
6477 51. Installed eximstats 1.31 in the distribution.
6479 52. Added support for SRV lookups to the dnslookup router.
6481 53. If an ACL referred to $message_body or $message_body_end, the value was not
6482 reset for any messages that followed in the same SMTP session.
6484 54. The store-handling optimization for building very long strings was not
6485 differentiating between the different store pools. I don't think this
6486 actually made any difference in practice, but I've tidied it.
6488 55. While running the routers to verify a sender address, $sender_address
6489 was still set to the sender address. This is wrong, because when routing to
6490 send a bounce to the sender, it would be empty. Therefore, I have changed
6491 it so that, while verifying a sender address, $sender_address is set to <>.
6492 (There is no change to what happens when verifying a recipient address.)
6494 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
6495 target A or AAAA records (if not already returned) without resetting the
6496 qualify_single or search_parents options of the DNS resolver. These are
6497 inappropriate in this case because the targets of MX and SRV records must
6498 be FQDNs. A broken DNS record could cause trouble if it happened to have a
6499 target that, when qualified, matched something in the local domain. These
6500 two options are now turned off when doing these lookups.
6502 57. It seems that at least some releases of Reiserfs (which does not have the
6503 concept of a fixed number of inodes) returns zero and not -1 for the
6504 number of available inodes. This interacted badly with check_spool_inodes,
6505 which assumed that -1 was the "no such thing" setting. What I have done is
6506 to check that the total number of inodes is greater than zero before doing
6507 the test of how many are available.
6509 58. When a "warn" ACL statement has a log_message modifier, the message is
6510 remembered, and not repeated. This is to avoid a lot of repetition when a
6511 message has many recipients that cause the same warning to be written.
6512 However, Exim was preserving the list of already written lines for an
6513 entire SMTP session, which doesn't seem right. The memory is now reset if a
6514 new message is started.
6516 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6517 debugging output unless log_rewrite was also set.
6519 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6520 of (int)(handle) when we know that handle contains (void *)(-1).
6522 61. The Exim daemon panic-logs an error return when it closes the incoming
6523 connection. However "connection reset by peer" seems to be common, and
6524 isn't really an error worthy of noting specially, so that particular error
6527 62. When Exim is trying to find all the local interfaces, it used to panic and
6528 die if the ioctl to get the interface flags failed. However, it seems that
6529 on at least one OS (Solaris 9) it is possible to have an interface that is
6530 included in the list of interfaces, but for which you get a failure error
6531 for this call. This happens when the interface is not "plumbed" into a
6532 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6533 failure of the "get flags" call assumes that the interface is down.
6535 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6536 makes life easier for people who are doing arithmetic on fields extracted
6537 from dates, where you often get leading zeros that should not be
6538 interpreted as octal.
6540 64. Added qualify_domain to the redirect router, to override the global
6543 65. If a pathologically long header line contained very many addresses (the
6544 report of this problem mentioned 10 000) and each of them was rewritten,
6545 Exim could use up a very large amount of memory. (It kept on making new
6546 copies of the header line as it rewrote, and never released the old ones.)
6547 At the expense of a bit more processing, the header rewriting function has
6548 been changed so that it no longer eats memory in this way.
6550 66. The generation of the Received: header has been moved from the time that a
6551 message starts to be received, to the time that it finishes. The timestamp
6552 in the Received: header should now be very close to that of the <= log
6553 line. There are two side-effects of this change:
6555 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6556 logged header lines no longer include the local Received: line, because
6557 it has not yet been created. The same applies to a copy of the message
6558 that is returned to a non-SMTP sender when a message is rejected.
6560 (b) When a filter file is tested using -bf, no additional Received: header
6561 is added to the test message. After some thought, I decided that this
6564 This change does not affect the value of $received_for. It is still set
6565 after address rewriting, but before local_scan() is called.
6567 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6569 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6570 gave an unhelpful panic error message, and a defer error. I have managed to
6571 change this behaviour so that it now rejects any supplied certificate,
6572 which seems right, as the list of acceptable certificates is empty.
6574 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6575 gave an unhelpful defer error. I have not managed to make this reject any
6576 supplied certificates, but the error message it gives is "no certificate
6577 supplied", which is not helpful.
6579 70. exigrep's output now also includes lines that are not associated with any
6580 message, but which match the given pattern. Implemented by a patch from
6581 Martin Sluka, which also tidied up the Perl a bit.
6583 71. Recipient callout verification, like sender verification, was using <> in
6584 the MAIL FROM command. This isn't really the right thing, since the actual
6585 sender may affect whether the remote host accepts the recipient or not. I
6586 have changed it to use the actual sender in the callout; this means that
6587 the cache record is now keyed on a recipient/sender pair, not just the
6588 recipient address. There doesn't seem to be a real danger of callout loops,
6589 since a callout by the remote host to check the sender would use <>.
6590 [SEE ABOVE: changed after hitting problems.]
6592 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6593 temporary errors. However, in the case of such a code being given after
6594 the end of a data transmission (i.e. after ".") Exim was failing to write
6595 a retry record for the message. (Yes, there was some broken host that was
6596 actually sending 8xx at this point.)
6598 73. An unknown lookup type in a host list could cause Exim to panic-die when
6599 the list was checked. (An example that provoked this was putting <; in the
6600 middle of a list instead of at the start.) If this happened during a DATA
6601 ACL check, a -D file could be left lying around. This kind of configuration
6602 error no longer causes Exim to die; instead it causes a defer error. The
6603 incident is still logged to the main and panic logs.
6605 74. Buglet left over from Exim 3 conversion. The message "too many messages
6606 in one connection" was written to the rejectlog but not the mainlog, except
6607 when address rewriting (yes!) was being logged.
6609 75. Added write_rejectlog option.
6611 76. When a system filter was run not as root (that is, when system_filter_user
6612 was set), the values of the $n variables were not being returned to the
6613 main process; thus, they were not subsequently available in the $sn
6616 77. Added +return_path_on_delivery log selector.
6618 78. A connection timeout was being treated differently from recipients deferred
6619 when testing hosts_max_try with a message that was older than the host's
6620 retry timeout. (The host should not be counted, thus allowing all hosts to
6621 be tried at least once before bouncing.) This may have been the cause of an
6622 occasionally reported bug whereby a message would remain on the queue
6623 longer than the retry timeout, but would be bounced if a delivery was
6624 forced. I say "may" because I never totally pinned down the problem;
6625 setting up timeout/retry tests is difficult. See also the next item.
6627 79. The ultimate address timeout was not being applied to errors that involved
6628 a combination of host plus message (for example, a timeout on a MAIL
6629 command). When an address resolved to a number of possible hosts, and they
6630 were not all tried for each delivery (e.g. because of hosts_max_try), a
6631 message could remain on the queue longer than the retry timeout.
6633 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6636 81. Fixed an obscure SMTP outgoing bug which required at least the following
6637 conditions: (a) there was another message waiting for the same server;
6638 (b) the server returned 5xx to all RCPT commands in the first message so
6639 that the message was not completed; (c) the server dropped the connection
6640 or gave a negative response to the RSET that Exim sends to abort the
6641 transaction. The observed case was a dropped connection after DATA that had
6642 been sent in pipelining mode. That is, the server had advertised PIPELINING
6643 but was not implementing it correctly. The effect of the bug was incorrect
6644 behaviour, such as trying another host, and this could lead to a crash.
6650 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6651 and daemon.c were passed as pointers to ints; they should have been
6652 pointers to socklen_t variables (which are typically unsigned ints).
6654 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6657 3. Fixed a really odd bug that affected only the testing scheme; patching a
6658 certain fixed string in the binary changed the value of another string that
6659 happened to be identical to the end of the original first string.
6661 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6662 name", it returns that address as the IP address. On some operating
6663 systems (e.g. Solaris), it also passes back the IP address string as the
6664 "host name". However, on others (e.g. Linux), it passes back an empty
6665 string. Exim wasn't checking for this, and was changing the host name to an
6666 empty string, assuming it had been canonicalized.
6668 5. Although rare, it is permitted to have more than one PTR record for a given
6669 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6670 all the names associated with an address, because they do in Solaris.
6671 However, it seems that they do not in Linux for data that comes from the
6672 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6673 I found this out when I moved to a new Linux workstation and tried to run
6674 the Exim test suite.
6676 To get round this problem I have changed the code so that it now does its
6677 own call to the DNS to look up PTR records when searching for a host name.
6678 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6679 addresses that are only in /etc/hosts are still found.
6681 This behaviour is, however, controlled by an option called host_lookup_
6682 order, which defaults to "bydns:byaddr". If people want to use the other
6683 order, or indeed, just use one or the other means of lookup, they can
6684 specify it in this variable.
6686 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6687 some operating systems, this comes back from gethostbyaddr() as an empty
6688 string, and this is what Exim used to test for. However, it seems that in
6689 other systems, "." is yielded. Exim now tests for this case too.
6691 7. The values of check_spool_space and check_log_space are now held internally
6692 as a number of kilobytes instead of an absolute number of bytes. If a
6693 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6694 kilobyte. This means that much larger values can be stored.
6696 8. Exim monitor: an attempt to get the action menu when not actually pointing
6697 at a message produces an empty menu entitled "No message selected". This
6698 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6699 no entries in it ("Shell widget menu has zero width and/or height"). So I
6700 have added a single, blank menu entry in this case.
6702 9. Added ${quote_local_part.
6704 10. MIME decoding is now applied to the contents of Subject: header lines when
6707 11. Now that a reference to $sender_host_address automatically causes a reverse
6708 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6709 host lookup before query-style lookups in lists that might use this
6710 variable. This has therefore been abolished, and the "net-" prefix is no
6711 longer necessary for query-style lookups.
6713 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6714 have been a typo for LFLAGS, so it has been changed.
6716 13. The install script calls Exim with "-C /dev/null" in order to find the
6717 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6718 to be output. However, since Exim outputs its version number before the
6719 error, it didn't break the script. It just looked ugly. I fixed this by
6720 always allowing "-C /dev/null" if the caller is root.
6722 14. Ignore overlarge ACL variable number when reading spool file - insurance
6723 against a later release with more variables having written the file.
6725 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6726 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6727 this, as well as the form without the "IPv6" on the front.
6729 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6730 OS/Makefile-Darwin file.
6732 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6733 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6735 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6736 variable. However, some OS use ints for inodes. I've added cast to long int
6737 to get rid of the compiler warning.
6739 19. I had forgotten to lock out "/../" in configuration file names when
6740 ALT_CONFIG_PREFIX was set.
6742 20. Routers used for verification do not need to specify transports. However,
6743 if such a router generated a host list, and callout was configured, Exim
6744 crashed, because it could not find a port number from the (non-existent)
6745 transport. It now assumes port 25 in this circumstance.
6747 21. Added the -t option to exigrep.
6749 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6750 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6751 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6752 them is set, LOOKUP_LSEARCH is forced.
6754 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6756 24. Added sender and host information to the "rejected by local_scan()" log
6757 line; previously there was no indication of these.
6759 25. Added .include_if_exists.
6761 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6762 Linux. It turns out that not all file systems support this. Apparently some
6763 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6764 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6765 not supported on the file descriptor, is now ignored when Exim is trying to
6766 sync a directory. This applies only to Linux.
6768 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6770 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6771 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6774 29. Added support for maildirsize files from supplied patch (modified a bit).
6776 30. The use of :fail: followed by an empty string could lead Exim to respond to
6777 sender verification failures with (e.g.):
6779 550 Verification failed for <xxx>
6780 550 Sender verify failed
6782 where the first response line was missing the '-' that indicates it is not
6783 the final line of the response.
6785 31. The loop for finding the name of the user that called Exim had a hardwired
6786 limit of 10; it now uses the value of finduser_retries, which is used for
6787 all other user lookups.
6789 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6792 33. Exim was neglecting to zero errno before one call of strtol() when
6793 expanding a string and expecting an integer value. On some systems this
6794 resulted in spurious "integer overflow" errors. Also, it was casting the
6795 result into an int without checking.
6797 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6798 did not work. The code looks as if it has *never* worked, though it appears
6799 to have been documented since at least release 1.62. I have made it work.
6801 35. The "timeout_DNS" error in retry rules, also documented since at least
6802 1.62, also never worked. As it isn't clear exactly what this means, and
6803 clearly it isn't a major issue, I have abolished the feature by treating it
6804 as "timeout", and writing a warning to the main and panic logs.
6806 36. The display of retry rules for -brt wasn't always showing the error code
6809 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6810 timeout_connect_A, timeout_connect_MX.
6812 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6813 to the empty sender.
6815 39. The daemon was not analysing the content of -oX till after it had closed
6816 stderr and disconnected from the controlling terminal. This meant that any
6817 syntax errors were only noted on the panic log, and the return code from
6818 the command was 0. By re-arranging the code a little, I've made the
6819 decoding happen first, so such errors now appear on stderr, and the return
6820 code is 1. However, the actual setting up of the sockets still happens in
6821 the disconnected process, so errors there are still only recorded on the
6824 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6825 connections (as happens on some IP stacks) was logged at start up time as
6826 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6827 it from "IPv6 and IPv4", which means that two separate sockets are being
6830 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6831 says whether AF_INET or AF_INET6 was passed as an argument.
6833 42. Exiwhat output was messed up when time zones were included in log
6836 43. Exiwhat now gives more information about the daemon's listening ports,
6837 and whether -tls-on-connect was used.
6839 44. The "port" option of the smtp transport is now expanded.
6841 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6842 silently ignored. Now an error message is written to the main and panic
6845 46. There's a new ACL modifier called "logwrite" which writes to a log file
6846 as soon as it is encountered.
6848 47. Added $local_user_uid and $local_user_gid at routing time.
6850 48. Exim crashed when trying to verify a sender address that was being
6853 49. Exim was recognizing only a space character after ".include". It now also
6854 recognizes a tab character.
6856 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6857 extracting the relevant information from the specification. The man page no
6858 longer contains scrambled data for the -d option, and I've added a section
6859 at the front about calling Exim under different names.
6861 51. Added "extra_headers" argument to the "mail" command in filter files.
6863 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6866 53. Installed eximstats 1.29.
6868 54. Added transport_filter_timeout as a generic transport option.
6870 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6871 Cc: header lines. This was required by RFC 822, but it not required by RFC
6874 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6875 incoming messages that did not have them. Now it does so only if the
6876 message originates locally, that is, if there is no associated remote host
6877 address. When Resent- header lines are present, this applies to the Resent-
6878 lines rather than the non-Resent- lines.
6880 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6881 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6883 58. Messages for configuration errors now include the name of the main
6884 configuration file - useful now that there may be more than one file in a
6885 list (.included file names were always shown).
6887 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6888 for those rare installations that do not start the daemon as root or run it
6889 setuid root. I've cut out the call to initgroups() if the daemon is not
6892 60. The Exim user and group can now be bound into the binary as text strings
6893 that are looked up at the start of Exim's processing.
6895 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6897 62. Added $mailstore_basename variable.
6899 63. Installed patch to sieve.c from Michael Haardt.
6901 64. When Exim failed to open the panic log after failing to open the main log,
6902 the original message it was trying to log was written to stderr and debug
6903 output, but if they were not available (the usual case in production), it
6904 was lost. Now it is written to syslog before the two lines that record the
6905 failures to open the logs.
6907 65. Users' Exim filters run in subprocesses under the user's uid. It is
6908 possible for a "deliver" command or an alias in a "personal" command to
6909 provoke an address rewrite. If logging of address rewriting is configured,
6910 this fails because the process is not running as root or exim. There may be
6911 a better way of dealing with this, but for the moment (because 4.30 needs
6912 to be released), I have disabled address rewrite logging when running a
6913 filter in a non-root, non-exim process.
6919 1. The buildconfig auxiliary program wasn't quoting the value set for
6920 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6921 not defined. This bug was masked in 4.22 by the effect that was fixed in
6924 2. Some messages that were rejected after a message id was allocated were
6925 shown as "incomplete" by exigrep. It no longer does this for messages that
6926 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6928 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6929 have allow_domain_literals set, the ID did not get logged in the <= line.
6930 Domain literals are now always recognized in Message-ID: header lines.
6932 4. The first argument for a ${extract expansion item is the key name or field
6933 number. Leading and trailing spaces in this item were not being ignored,
6934 causing some misleading effects.
6936 5. When deliver_drop_privilege was set, single queue runner processes started
6937 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6938 same command in the process it spins off) were not dropping privilege.
6940 6. When the daemon running as "exim" started a queue runner, it always
6941 re-executed Exim in the spun-off process. This is a waste of effort when
6942 deliver_drop_privilege is set. The new process now just calls the
6943 queue-runner function directly.
6949 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6952 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6953 ignored. Though the use of -oP was forcing the writing of a pid file, it
6954 was always written to the default place.
6956 3. If the message "no IP address found for host xxxx" is generated during
6957 incoming verification, it is now followed by identification of the incoming
6958 connection (so you can more easily find what provoked it).
6960 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6962 5. Added some features to "harden" Exim a bit more against certain attacks:
6964 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6965 be put in Local/Makefile. This is like the never_users runtime option,
6966 but it cannot be overridden. The default setting is "root".
6968 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6969 prefix string with which any file named in a -C command line option
6972 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6973 is retained for -C and -D only if the caller of Exim is root. Without
6974 it, the exim user may also use -C and -D and retain privilege.
6976 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6977 command line option is disabled.
6979 6. Macro names set by the -D option must start with an upper case letter, just
6980 like macro names defined in the configuration file.
6982 7. Added "dereference=" facility to LDAP.
6984 8. Two instances of the typo "uknown" in the source files are fixed.
6986 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6987 the Configure-Makefile script screwed up while processing it.
6989 10. Incorporated PCRE 4.4.
6991 11. The SMTP synchronization check was not operating right at the start of an
6992 SMTP session. For example, it could not catch a HELO sent before the client
6993 waited for the greeting. There is now a check for outstanding input at the
6994 point when the greeting is written. Because of the duplex, asynchronous
6995 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6996 way, but not yet received, when the check is performed.
6998 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6999 on TCP/IP sockets, because this apparently causes some broken clients to
7002 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
7003 unchanged) from the Cygwin maintainer.
7005 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
7006 was supported instead of testing for mbx. Effectively a typo.
7008 15. The spa authenticator server code was not checking that the input it
7009 received was valid base64.
7011 16. The debug output line for the "set" modifier in ACLs was not showing the
7012 name of the variable that was being set.
7014 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
7016 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
7017 lookup on its own. Something else was needed to trigger the lookup. For
7018 example, a match in host_lookup or the need for a host name in a host list.
7019 Now, if $sender_host_name is referenced and the host name has not yet been
7020 looked up, a lookup is performed. If the lookup fails, the variable remains
7021 empty, and $host_lookup_failed is set to "1".
7023 19. Added "eqi" as a case-independent comparison operator.
7025 20. The saslauthd authentication condition could segfault if neither service
7026 nor realm was specified.
7028 21. If an overflowing value such as "2048M" was set for message_size_limit, the
7029 error message that was logged was misleading, and incoming SMTP
7030 connections were dropped. The message is now more accurate, and temporary
7031 errors are given to SMTP connections.
7033 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
7034 (except RSET) with a 421 error, until QUIT is received. However, it was
7035 failing to send a response to QUIT.
7037 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
7038 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
7039 ACL is now run after the helo_try_verify_hosts code.
7041 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
7042 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
7043 case-independent, so other case variants are also recognized). Apparently
7044 some systems use these upper case variants.
7046 25. If more than two messages were waiting for the same host, and a transport
7047 filter was specified for the transport, Exim sent two messages over the
7048 same TCP/IP connection, and then failed with "socket operation on non-
7049 socket" when it tried to send the third.
7051 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
7053 27. The extern definition of crypt16() in expand.c was not being excluded when
7054 the OS had its own crypt16() function.
7056 28. Added bounce_return_body as a new option, and bounce_return_size_limit
7057 as a preferred synonym for return_size_limit, both as an option and as an
7060 29. Added LIBS=-liconv to OS/Makefile-OSF1.
7062 30. Changed the default configuration ACL to relax the local part checking rule
7063 for addresses that are not in any local domains. For these addresses,
7064 slashes and pipe symbols are allowed within local parts, but the sequence
7065 /../ is explicitly forbidden.
7067 31. SPA server authentication was not clearing the challenge buffer before
7070 32. log_message in a "warn" ACL statement was writing to the reject log as
7071 well as to the main log, which contradicts the documentation and doesn't
7072 seem right (because no rejection is happening). So I have stopped it.
7074 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
7075 However, I am unable to do any testing of this.
7077 34. Fixed an infelicity in the appendfile transport. When checking directories
7078 for a mailbox, to see if any needed to be created, it was accidentally
7079 using path names with one or more superfluous leading slashes; tracing
7080 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
7082 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
7083 contents are added to the log line that is written for every discarded
7084 recipient. (Previously a log_message setting was ignored.)
7086 36. The ${quote: operator now quotes the string if it is empty.
7088 37. The install script runs exim in order to find its version number. If for
7089 some reason other than non-existence or emptiness, which it checks, it
7090 could not run './exim', it was installing it with an empty version number,
7091 i.e. as "exim-". This error state is now caught, and the installation is
7094 38. An argument was missing from the function that creates an error message
7095 when Exim fails to connect to the socket for saslauthd authentication.
7096 This could cause Exim to crash, or give a corrupted message.
7098 39. Added isip, isip4, and isip6 to ${if conditions.
7100 40. The ACL variables $acl_xx are now saved with the message, and can be
7101 accessed later in routers, transports, and filters.
7103 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
7104 strings in the file are not string-expanded.
7106 42. If a MAIL command specified a SIZE value that was too large to fit into an
7107 int variable, the check against message_size_limit failed. Such values are
7108 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
7109 day this will have to be increased, but I don't think I want to be around
7110 when emails are that large.
7117 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
7118 iconv() is not standard in FreeBSD.
7120 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
7121 IPv6 enabled. The observed symptom was a segmentation fault on return from
7122 the function os_common_find_running_interfaces() in src/os.c.
7124 3. In the check_special_case() function in daemon.c I had used "errno" as an
7125 argument name, which causes warnings on some systems. This was basically a
7126 typo, since it was named "eno" in the comments!
7128 4. The code that waits for the clock to tick (at a resolution of some fraction
7129 of a second) so as to ensure message-id uniqueness was always waiting for
7130 at least one whole tick, when it could have waited for less. [This is
7131 almost certainly not relevant at current processor speeds, where it is
7132 unlikely to ever wait at all. But we try to future-proof.]
7134 5. The function that sleeps for a time interval that includes fractions of a
7135 second contained a race. It did not block SIGALRM between setting the
7136 timer, and suspending (a couple of lines later). If the interval was short
7137 and the sigsuspend() was delayed until after it had expired, the suspension
7138 never ended. On busy systems this could lead to processes getting stuck for
7141 6. Some uncommon configurations may cause a lookup to happen in a queue runner
7142 process, before it forks any delivery processes. The open lookup caching
7143 mechanism meant that the open file or database connection was passed into
7144 the delivery process. The problem was that delivery processes always tidy
7145 up cached lookup data. This could cause a problem for the next delivery
7146 process started by the queue runner, because the external queue runner
7147 process does not know about the closure. So the next delivery process
7148 still has data in the lookup cache. In the case of a file lookup, there was
7149 no problem because closing a file descriptor in a subprocess doesn't affect
7150 the parent. However, if the lookup was caching a connection to a database,
7151 the connection was closed, and the second delivery process was likely to
7152 see errors such as "PGSQL: query failed: server closed the connection
7153 unexpectedly". The problem has been fixed by closing all cached lookups
7154 in a queue runner before running a delivery process.
7156 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
7157 seem to have the "const" qualifier which it has on other OS. I've
7160 8. Change 4.21/2 was too strict. It is only if there are two authenticators
7161 *of the same type* (client or server) with the same public name that an
7162 error should be diagnosed.
7164 9. When Exim looked up a host name for an IP address, but failed to find the
7165 original IP address when looking up the host name (a safety check), it
7166 output the message "<ip address> does not match any IP for NULL", which was
7167 confusing, to say the least. The bug was that the host name should have
7168 appeared instead of "NULL".
7170 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
7171 user that is built into the binary, and the -C or -D options is used, root
7172 privilege is dropped before the configuration file is read. In addition,
7173 logging is switched to stderr instead of the normal log files. If the
7174 configuration then re-defines the Exim user, the unprivileged environment
7175 is probably not what is expected, so Exim logs a panic warning message (but
7178 However, if deliver_drop_privilege is set, the unprivileged state may well
7179 be exactly what is intended, so the warning has been cut out in that case,
7180 and Exim is allowed to try to write to its normal log files.
7186 1. smtp_return_error_details was not giving details for temporary sender
7187 or receiver verification errors.
7189 2. Diagnose a configuration error if two authenticators have the same public
7192 3. Exim used not to create the message log file for a message until the first
7193 delivery attempt. This could be confusing when incoming messages were held
7194 for policy or load reasons. The message log file is now created at the time
7195 the message is received, and an initial "Received" line is written to it.
7197 4. The automatically generated man page for command line options had a minor
7198 bug that caused no ill effects; however, a more serious problem was that
7199 the procedure for building the man page automatically didn't always
7200 operate. Consequently, release 4.20 contains an out-of-date version. This
7201 shouldn't happen again.
7203 5. When building Exim with embedded Perl support, the script that builds the
7204 Makefile was calling 'perl' to find its compile-time parameters, ignoring
7205 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
7207 6. The freeze_tell option was not being used for messages that were frozen on
7208 arrival, either by an ACL or by local_scan().
7210 7. Added the smtp_incomplete_transaction log selector.
7212 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
7213 was accepting AUTH without a new EHLO.
7215 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
7216 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
7219 10. Logging of TCP/IP connections (when configured) now happens in the main
7220 daemon process instead of the child process, so that the TCP/IP connection
7221 count is more accurate (but it can never be perfect).
7223 11. The use of "drop" in a nested ACL was not being handled correctly in the
7224 outer ACL. Now, if condition failure induced by the nested "drop" causes
7225 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
7226 or "require"), the connection is dropped.
7228 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
7229 that yield "discard" can now be used with an "accept" or a "discard" verb,
7230 but an error is generated for any others (because I can't see a useful way
7231 to define what should happen).
7233 13. When an ACL is read dynamically from a file (or anywhere else), the lines
7234 are now processed in the same way as lines in the Exim configuration file.
7235 In particular, continuation lines are supported.
7237 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
7239 15. Added -ti meaning -t -i.
7241 16. Check for letters, digits, hyphens, and dots in the names of dnslist
7242 domains, and warn by logging if others are found.
7244 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
7245 returned from GIFCONF when Exim is trying to find the list of interfaces on
7246 a host. The code in os.c has been modified to copy each ifreq to an aligned
7247 structure in all cases.
7249 Also, in some cases, the returned ifreq's were being copied to a 'struct
7250 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
7251 means the last couple of bytes of an IPv6 address could be chopped if the
7252 ifreq contained only a normal sockaddr (14 bytes storage).
7254 18. Named domain lists were not supported in the hosts_treat_as_local option.
7255 An entry such as +xxxx was not recognized, and was treated as a literal
7258 19. Ensure that header lines added by a DATA ACL are included in the reject log
7259 if the ACL subsequently rejects the message.
7261 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
7262 MD5 (which is deprecated).
7264 21. When testing a filter file using -bf, Exim was writing a message when it
7265 took the sender from a "From " line in the message, but it was not doing so
7266 when it took $return_path from a Return-Path: header line. It now does.
7268 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
7269 with a valid header line field name (a series of printing characters
7270 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
7272 23. Changed "disc" in the source to "disk" to conform to the documentation and
7273 the book and for uniformity.
7275 24. Ignore Sendmail's -Ooption=value command line item.
7277 25. When execve() failed while trying to run a command in a pipe transport,
7278 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
7279 could be confused with a return value of 69 from the command itself. This
7280 has been changed to 127, the value the shell returns if it is asked to run
7281 a non-existent command. The wording for the related log line suggests a
7282 non-existent command as the problem.
7284 26. If received_header_text expands to an empty string, do not add a Received:
7285 header line to the message. (Well, it adds a token one on the spool, but
7286 marks it "old" so that it doesn't get used or transmitted.)
7288 27. Installed eximstats 1.28 (addition of -nt option).
7290 28. There was no check for failure on the call to getsockname() in the daemon
7291 code. This can fail if there is a shortage of resources on the system, with
7292 ENOMEM, for example. A temporary error is now given on failure.
7294 29. Contrary to the C standard, it seems that in some environments, the
7295 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
7296 program. Exim now does this explicitly; it affects the formatting of
7297 timestamps using strftime().
7299 30. If exiqsumm was given junk data, it threw up some uninitialized variable
7300 complaints. I've now initialized all the variables, to avoid this.
7302 32. Header lines added by a system filter were not being "seen" during
7303 transport-time rewrites.
7305 33. The info_callback() function passed to OpenSSL is set up with type void
7306 (*)(SSL *, int, int), as described somewhere. However, when calling the
7307 function (actually a macro) that sets it up, the type void(*)() is
7308 expected. I've put in a cast to prevent warnings from picky compilers.
7310 34. If a DNS black list lookup found a CNAME record, but there were no A
7311 records associated with the domain it pointed at, Exim crashed.
7313 35. If a DNS black list lookup returned more than one A record, Exim ignored
7314 all but the first. It now scans all returned addresses if a particular IP
7315 value is being sought. In this situation, the contents of the
7316 $dnslist_value variable are a list of all the addresses, separated by a
7319 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
7320 to accept a host name and all its aliases if the forward lookup for any of
7321 them yielded the IP address of the incoming connection. Now it accepts only
7322 those names whose forward lookup yields the correct IP address. Any other
7323 names are discarded. This closes a loophole whereby a rogue DNS
7324 administrator could create reverse DNS records to break through a
7325 wildcarded host restriction in an ACL.
7327 37. If a user filter or a system filter that ran in a subprocess used any of
7328 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
7329 the wrong values were passed to the pipe command ($thisaddress had the
7330 value of $0, $0 had the value of $1, etc). This bug was introduced by
7331 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
7333 38. Improved the line breaking for long SMTP error messages from ACLs.
7334 Previously, if there was no break point between 40 and 75 characters, Exim
7335 left the rest of the message alone. Two changes have been made: (a) I've
7336 reduced the minimum length to 35 characters; (b) if it can't find a break
7337 point between 35 and 75 characters, it looks ahead and uses the first one
7338 that it finds. This may give the occasional overlong line, but at least the
7339 remaining text gets split now.
7341 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
7342 file descriptors might be low, and that setting 1000 would always raise it.
7343 It turns out that in some environments, the limit is already over 1000 and
7344 that lowering it causes trouble. So now Exim takes care not to decrease it.
7346 40. When delivering a message, the value of $return_path is set to $sender_
7347 address at the start of routing (routers may change the value). By an
7348 oversight, this default was not being set up when an address was tested by
7349 -bt or -bv, which affected the outcome if any router or filter referred to
7352 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
7353 log only when "message" or "log_message" are set. However, if one of the
7354 conditions was an address verification, or a call to a nested ACL, the
7355 messages generated by the underlying test were being passed through. This
7356 no longer happens. The underlying message is available in $acl_verify_
7357 message for both "message" and "log_message" expansions, so it can be
7358 passed through if needed.
7360 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
7361 new expansion $bh_ to give the encoded byte string without charset
7362 translation. Translation happens only if iconv() is available; HAVE_ICONV
7363 indicates this at build time. HEADERS_CHARSET gives the charset to
7364 translate to; headers_charset can change it in the configuration, and
7365 "headers charset" can change it in an individual filter file.
7367 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
7368 that creates RFC 2047 encoded "words" labels them as that charset instead
7369 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
7370 expansion operator; (ii) when Exim creates a From: line for a local
7371 message; (iii) when a header line is rewritten to include a "phrase" part.
7373 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
7374 buggy, causing it to skip the first lines of messages whose message ID
7375 ended in 'D'. This would not have bitten before Exim release 4.14, because
7376 message IDs were unlikely to end in 'D' before then. The effect was to have
7377 incorrect size information for certain domains.
7379 45. #include "config.h" was missing at the start of the crypt16.c module. This
7380 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
7383 46. If there was a timeout during a "random" callout check, Exim treated it as
7384 a failure of the random address, and carried on sending RSET and the real
7385 address. If the delay was just some slowness somewhere, the response to the
7386 original RCPT would be taken as a response to RSET and so on, causing
7387 mayhem of various kinds.
7389 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
7390 when I implemented it. It didn't allow for the fact that some option values
7391 may legitimately be negative (e.g. size_addition), and it didn't even do
7392 the right test for positive values.
7394 48. Domain names in DNS records are case-independent. Exim always looks them up
7395 in lower case. Some resolvers return domain names in exactly the case they
7396 appear in the zone file, that is, they may contain uppercase letters. Not
7397 all resolvers do this - some return always lower case. Exim was treating a
7398 change of case by a resolver as a change of domain, similar to a widening
7399 of a domain abbreviation. This triggered its re-routing code and so it was
7400 trying to route what was effectively the same domain again. This normally
7401 caused routing to fail (because the router wouldn't handle the domain
7402 twice). Now Exim checks for this case specially, and just changes the
7403 casing of the domain that it ultimately uses when it transmits the message
7406 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
7409 50. If a filter generated a file delivery with a non-absolute name (possible if
7410 no home directory exists for the router), the forbid_file option was not
7413 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
7414 the existing equality matching.
7416 52. Exim was using ints instead of ino_t variables in some places where it was
7417 dealing with inode numbers.
7419 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
7420 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
7421 called TMPDIR, and if it finds it is different, it changes its value.
7423 54. The smtp_printf() function is now made available to local_scan() so
7424 additional output lines can be written before returning. There is also an
7425 smtp_fflush() function to enable the detection of a dropped connection.
7426 The variables smtp_input and smtp_batched_input are exported to
7429 55. Changed the default runtime configuration: the message "Unknown user"
7430 has been removed from the ACL, and instead placed on the localuser router,
7431 using the cannot_route_message feature. This means that any verification
7432 failures that generate their own messages won't get overridden. Similarly,
7433 the "Unrouteable address" message that was in the ACL for unverifiable
7434 relay addresses has also been removed.
7436 56. Added hosts_avoid_esmtp to the smtp transport.
7438 57. The exicyclog script was not checking for the esoteric option
7439 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
7440 will work only if exicyclog is run under the appropriate euid.
7442 58. Following a discussion on the list, the rules by which Exim recognises line
7443 endings on incoming messages have been changed. The -dropcr and drop_cr
7444 options are now no-ops, retained only for backwards compatibility. The
7445 following line terminators are recognized: LF CRLF CR. However, special
7446 processing applies to CR:
7448 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
7449 nor a local message in the state where . is a terminator.
7451 (ii) If a bare CR is encountered in a header line, an extra space is added
7452 after the line terminator so as not to end the header. The reasoning
7453 behind this is that bare CRs in header lines are most likely either
7454 to be mistakes, or people trying to play silly games.
7456 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
7457 was being incorrectly given as 18 bytes larger than it should have been.
7458 This is a VOB (very old bug).
7460 60. This may never have affected anything current, but just in case it has:
7461 When the local host is found other than at the start of a list of hosts,
7462 the local host, those with the same MX, and any that follow, are discarded.
7463 When the list in question was part of a longer list of hosts, the following
7464 hosts (not currently being processed) were also being discarded. This no
7465 longer happens. I'm not sure if this situation could ever has previously
7468 61. Added the "/MX" feature to lists of hosts in the manualroute and query
7471 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
7472 header. This is something that is recommended in a new Internet Draft, and
7473 is something that is documented as being done by Sendmail. There are two
7474 possible values. For messages generated by the autoreply transport, Exim
7477 Auto-Submitted: auto-replied
7479 whereas for all other generated messages (e.g. bounces) it adds
7481 Auto-Submitted: auto-generated
7483 63. The "personal" condition in filters now includes a test for the
7484 Auto-Submitted: header. If it contains the string "auto-" the message it
7485 not considered personal.
7487 64. Added rcpt_include_affixes as a generic transport option.
7489 65. Added queue_only_override (default true).
7491 66. Added the syslog_duplication option.
7493 67. If what should have been the first header line of a message consisted of
7494 a space followed by a colon, Exim was mis-interpreting it as a header line.
7495 It isn't of course - it is syntactically invalid and should therefore be
7496 treated as the start of the message body. The misbehaviour could have
7497 caused a number of strange effects, including loss of data in subsequent
7498 header lines, and spool format errors.
7500 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
7501 client host had authenticated. This control can now be exercised by an ACL
7502 for more flexibility.
7504 69. By default, callouts do not happen when testing with -bh. There is now a
7505 variant, -bhc, which does actually run the callout code, including
7506 consulting and updating the callout cache.
7508 70. Added support for saslauthd authentication, courtesy of Alexander
7511 71. If statvfs() failed on the spool or log directories while checking their
7512 size for availability, Exim confusingly gave the error "space shortage".
7513 Furthermore, in debugging mode it crashed with a floating point exception.
7514 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7515 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7516 serious problem, Exim now writes to the main and panic logs when this
7517 happens, with details of the failure. It then refuses to accept the
7518 incoming message, giving the message "spool directory problem" or "log
7519 directory problem" with a 421 code for SMTP messages.
7521 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7522 0, 1, and 2 exist, because some OS complain for execs without them (see
7523 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7524 descriptors. However, the code omitted to check that the open succeeded,
7525 causing mysterious errors if for some reason the permissions on /dev/null
7526 got screwed. Now Exim writes a message to the main and panic logs, and
7527 bombs out if it can't open /dev/null.
7529 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7530 interact so that it is all more flexible. It is supposed to remain
7531 backwards compatible. Also added extra_local_interfaces.
7533 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7534 to bomb out with an assertion failure - to the client this appears as a
7535 connection drop. This problem occurs in the part of the code that was taken
7536 from the Samba project. Fortunately, the assertion is in a very simple
7537 function, so I have fixed this by reproducing the function inline in the
7538 one place where it is called, and arranging for authentication to fail
7539 instead of killing the process with assert().
7541 75. The SPA client code was not working when the server requested OEM rather
7542 than Unicode encoding.
7544 76. Added code to make require_files with a specific uid setting more usable in
7545 the case where statting the file as root fails - usually a non-root-mounted
7546 NFS file system. When this happens and the failure is EACCES, Exim now
7547 forks a subprocess and does the per-uid checking as the relevant uid.
7549 77. Added process_log_path.
7551 78. If log_file_path was not explicitly set, a setting of check_log_space or
7552 check_log_inodes was ignored.
7554 79. If a space check for the spool or log partitions fails, the incident is now
7555 logged. Of course, in the latter case the data may get lost...
7557 80. Added the %p formatting code to string_format() so that it can be used to
7558 print addresses in debug_print(). Adjusted all the address printing in the
7559 debugging in store.c to use %p rather than %d.
7561 81. There was a concern that a line of code in smtp_in.c could overflow a
7562 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7563 initially expressed, the concern was not well-founded, because trailing
7564 spaces are removed early. However, if the trailing spaces were followed by
7565 a NULL, they did not get removed, so the overflow was possible. Two fixes
7568 (a) I re-wrote the offending code in a cleaner fashion.
7569 (b) If an incoming SMTP command contains a NULL character, it is rejected
7572 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7573 runs initgroups(), so that if the Exim user is in any additional groups,
7574 they will be used during message reception.
7580 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob