1 This document describes *changes* to previous versions, that might
2 affect Exim's operation, with an unchanged configuration file. For new
3 options, and new features, see the NewStuff file next to this ChangeLog.
9 JH/01 OpenSSL: With debug enabled output keying information sufficient, server
10 side, to decode a TLS 1.3 packet capture.
12 JH/02 OpenSSL: Suppress the sending of (stateful) TLS1.3 session tickets.
13 Previously the default library behaviour applied, sending two, each in
16 JH/03 Debug output for ACL now gives the config file name and line number for
19 JH/04 The default received_header_text now uses the RFC 8314 tls cipher clause.
21 JH/05 DKIM: ensure that dkim_domain elements are lowercased before use.
23 JH/06 Fix buggy handling of autoreply bounce_return_size_limit, and a possible
24 buffer overrun for (non-chunking) other transports.
26 JH/07 GnuTLS: Our use of late (post-handshake) certificate verification, under
27 TLS1.3, means that a server rejecting a client certificate is not visible
28 to the client until the first read of encrypted data (typically the
29 response to EHLO). Add detection for that case and treat it as a failed
30 TLS connection attempt, so that the normal retry-in-clear can work (if
33 JB/01 Bug 2375: fix expansions of 822 addresses having comments in local-part
34 and/or domain. Found and fixed by Jason Betts.
36 JH/08 Add hardening against SRV & TLSA lookups the hit CNAMEs (a nonvalid
37 configuration). If a CNAME target was not a wellformed name pattern, a
40 JH/09 Logging: Fix initial listening-on line for multiple ports for an IP when
41 the OS reports them interleaved with other addresses.
43 JH/10 OpenSSL: Fix aggregation of messages. Previously, when PIPELINING was
44 used both for input and for a verify callout, both encrypted, SMTP
45 responses being sent by the server could be lost. This resulted in
46 dropped connections and sometimes bounces generated by a peer sending
49 JH/11 Harden plaintext authenticator against a badly misconfigured client-send
50 string. Previously it was possible to cause undefined behaviour in a
51 library routine (usually a crash). Found by "zerons".
53 JH/12 Bug 2384: fix "-bP smtp_receive_timeout". Previously it returned no
56 JH/13 Bug 2386: Fix builds with Dane under LibreSSL 2.9.0 onward. Some old
57 API was removed, so update to use the newer ones.
59 JH/14 Bug 1891: Close the log file if receiving a non-smtp message, without
60 any timeout set, is taking a long time. Previously we would hang on to a
61 rotated logfile "forever" if the input was arriving with long gaps
62 (a previous attempt to fix addressed lack, for a long time, of initial
65 HS/01 Bug 2390: Use message_id for tempfile creation to avoid races in a
66 shared (NFS) environment. The length of the tempfile name is now
67 4 + 16 ("hdr.$message_exim_id") which might break on file
68 systems which restrict the file name length to lower values.
71 HS/02 Bug 2390: Use message_id for tempfile creation to avoid races in a
72 shared (NFS) environment.
74 HS/03 Bug 2392: exigrep does case sensitive *option* processing (as it
75 did for all versions <4.90). Notably -M, -m, --invert, -I may be
78 JH/15 Use unsigned when creating bitmasks in macros, to avoid build errors
79 on some platforms for bit 31.
81 JH/16 GnuTLS: rework ciphersuite strings under recent library versions. Thanks
82 to changes apparently associated with TLS1.3 handling some of the APIs
83 previously used were either nonfunctional or inappropriate. Strings
84 like TLS1.3:ECDHE_SECP256R1__RSA_PSS_RSAE_SHA256__AES_256_GCM__AEAD:256
85 and TLS1.2:ECDHE_SECP256R1__RSA_SHA256__AES_128_CBC__SHA256:128 replace
86 the previous TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256 .
87 This affects log line X= elements, the $tls_{in,out}_cipher variables,
88 and the use of specific cipher names in the encrypted= ACL condition.
90 JH/17 OpenSSL: the default openssl_options now disables ssl_v3.
92 JH/18 GnuTLS: fix $tls_out_ocsp under hosts_request_ocsp. Previously the
93 verification result was not updated unless hosts_require_ocsp applied.
95 JH/19 Bug 2398: fix listing of a named-queue. Previously, even with the option
96 queue_list_requires_admin set to false, non-admin users were denied the
99 JH/20 Bug 2389: fix server advertising of usable certificates, under GnuTLS in
100 directory-of-certs mode. Previously they were advertised despite the
103 JH/21 The smtp transport option "hosts_noproxy_tls" is now unset by default.
104 A single TCP connection by a client will now hold a TLS connection open
105 for multiple message deliveries, by default. Previoud the default was to
108 JH/22 The smtp transport option "hosts_try_dane" now enables all hosts by
109 default. If built with the facility, DANE will be used. The facility
110 SUPPORT_DANE is now enabled in the prototype build Makefile "EDITME".
112 JH/23 The build default is now for TLS to be included; the SUPPORT_TLS define
113 is replaced with DISABLE_TLS. Either USE_GNUTLS or (the new) USE_OPENSSL
114 must be defined and you must still, unless you define DISABLE_TLS, manage
115 the the include-dir and library-file requirements that go with that
116 choice. Non-TLS builds are still supported.
118 JH/24 Fix duplicated logging of peer name/address, on a transport connection-
121 JH/25 The smtp transport option "hosts_try_fastopen" now enables all hosts by
122 default. If the platform supports and has the facility enabled, it will
123 be requested on all coneections.
125 JH/26 The PIPE_CONNECT facility is promoted from experimental status and is now
126 controlled by the build-time option SUPPORT_PIPE_CONNECT.
128 PP/01 Unbreak heimdal_gssapi, broken in 4.92.
130 JH/27 Bug 2404: Use the main-section configuration option "dsn_from" for
131 success-DSN messages. Previously the From: header was always the default
132 one for these; the option was ignored.
134 JH/28 Fix the timeout on smtp response to apply to the whole response.
135 Previously it was reset for every read, so a teergrubing peer sending
136 single bytes within the time limit could extend the connection for a
137 long time. Credit to Qualsys Security Advisory Team for the discovery.
139 JH/29 Fix DSN Final-Recipient: field. Previously it was the post-routing
140 delivery address, which leaked information of the results of local
141 forwarding. Change to the original envelope recipient address, per
144 JH/30 Bug 2411: Fix DSN generation when RFC 3461 failure notification is
145 requested. Previously not bounce was generated and a log entry of
146 error ignored was made.
148 JH/31 Avoid re-expansion in ${sort } expansion. (CVE-2019-13917)
150 JH/32 Introduce a general tainting mechanism for values read from the input
151 channel, and values derived from them. Refuse to expand any tainted
152 values, to catch one form of exploit.
154 JH/33 Bug 2413: Fix dkim_strict option. Previously the expansion result
155 was unused and the unexpanded text used for the test. Found and
156 fixed by Ruben Jenster.
158 JH/34 Fix crash after TLS shutdown. When the TCP/SMTP channel was left open,
159 an attempt to use a TLS library read routine dereffed a nul pointer,
162 JH/35 Bug 2409: filter out-of-spec chars from callout response before using
163 them in our smtp response.
165 JH/36 Have the general router option retry_use_local_part default to true when
166 any of the restrictive preconditions are set (to anything). Previously it
167 was only for check_local user. The change removes one item of manual
168 configuration which is required for proper retries when a remote router
169 handles a subset of addresses for a domain.
171 JH/37 Appendfile: when evaluating quota use (non-quota_size_regex) take the file
172 link count into consideration.
174 HS/04 Fix handling of very log lines in -H files. If a -<key> <value> line
175 caused the extension of big_buffer, the following lines were ignored.
177 JH/38 Bug 1395: Teach the DNS negative-cache about TTL value from the SOA in
178 accordance with RFC 2308. Previously there was no expiry, so a longlived
179 receive process (eg. due to ACL delays) versus a short SOA value could
182 HS/05 Handle trailing backslash gracefully. (CVE-2019-15846)
184 JH/39 Promote DMARC support to mainline.
186 JH/40 Bug 2452: Add a References: header to DSNs.
188 JH/41 With GnuTLS 3.6.0 (and later) do not attempt to manage Diffie-Hellman
189 parameters. The relevant library call is documented as "Deprecated: This
190 function is unnecessary and discouraged on GnuTLS 3.6.0 or later. Since
191 3.6.0, DH parameters are negotiated following RFC7919."
193 HS/06 Change the default of dnssec_request_domains to "*"
195 JH/42 Bug 2545: Fix CHUNKING for all RCPT commands rejected. Previously we
196 carried on and emitted a BDAT command, even when PIPELINING was not
199 JH/43 Bug 2465: Fix taint-handling in dsearch lookup. Previously a nontainted
200 buffer was used for the filename, resulting in a trap when tainted
201 arguments (eg. $domain) were used.
203 JH/44 With OpenSSL 1.1.1 (onwards) disable renegotiation for TLS1.2 and below;
204 recommended to avoid a possible server-load attack. The feature can be
205 re-enabled via the openssl_options main cofiguration option.
207 JH/45 local_scan API: documented the current smtp_printf() call. This changed
208 for version 4.90 - adding a "more data" boolean to the arguments.
209 Bumped the ABI version number also, this having been missed previously;
210 release versions 4.90 to 4.92.3 inclusive were effectively broken in
211 respect of usage of smtp_printf() by either local_scan code or libraries
212 accessed via the ${dlfunc } expansion item. Both will need coding
213 adjustment for any calls to smtp_printf() to match the new function
214 signature; a FALSE value for the new argument is always safe.
220 JH/01 Remove code calling the customisable local_scan function, unless a new
221 definition "HAVE_LOCAL_SCAN=yes" is present in the Local/Makefile.
223 JH/02 Bug 1007: Avoid doing logging from signal-handlers, as that can result in
224 non-signal-safe functions being used.
226 JH/03 Bug 2269: When presented with a received message having a stupidly large
227 number of DKIM-Signature headers, disable DKIM verification to avoid
228 a resource-consumption attack. The limit is set at twenty.
230 JH/04 Add variables $arc_domains, $arc_oldest_pass for ARC verify. Fix the
231 report of oldest_pass in ${authres } in consequence, and separate out
232 some descriptions of reasons for verification fail.
234 JH/05 Bug 2273: Cutthrough delivery left a window where the received messsage
235 files in the spool were present and unlocked. A queue-runner could spot
236 them, resulting in a duplicate delivery. Fix that by doing the unlock
237 after the unlink. Investigation by Tim Stewart. Take the opportunity to
238 add more error-checking on spoolfile handling while that code is being
241 PP/01 Refuse to open a spool data file (*-D) if it's a symlink.
242 No known attacks, no CVE, this is defensive hardening.
244 JH/06 Bug 2275: The MIME ACL unlocked the received message files early, and
245 a queue-runner could start a delivery while other operations were ongoing.
246 Cutthrough delivery was a common victim, resulting in duplicate delivery.
247 Found and investigated by Tim Stewart. Fix by using the open message data
248 file handle rather than opening another, and not locally closing it (which
249 releases a lock) for that case, while creating the temporary .eml format
250 file for the MIME ACL. Also applies to "regex" and "spam" ACL conditions.
252 JH/07 Bug 177: Make a random-recipient callout success visible in ACL, by setting
253 $sender_verify_failure/$recipient_verify_failure to "random".
255 JH/08 When generating a selfsigned cert, use serial number 1 since zero is not
258 JH/09 Bug 2274: Fix logging of cmdline args when starting in an unlinked cwd.
259 Previously this would segfault.
261 JH/10 Fix ARC signing for case when DKIM signing failed. Previously this would
264 JH/11 Bug 2264: Exim now only follows CNAME chains one step by default. We'd
265 like zero, since the resolver should be doing this for us, But we need one
266 as a CNAME but no MX presence gets the CNAME returned; we need to check
267 that doesn't point to an MX to declare it "no MX returned" rather than
268 "error, loop". A new main option is added so the older capability of
269 following some limited number of chain links is maintained.
271 JH/12 Add client-ip info to non-pass iprev ${authres } lines.
273 JH/13 For receent Openssl versions (1.1 onward) use modern generic protocol
274 methods. These should support TLS 1.3; they arrived with TLS 1.3 and the
275 now-deprecated earlier definitions used only specified the range up to TLS
276 1.2 (in the older-version library docs).
278 JH/14 Bug 2284: Fix DKIM signing for body lines starting with a pair of dots.
280 JH/15 Rework TLS client-side context management. Stop using a global, and
281 explicitly pass a context around. This enables future use of TLS for
282 connections to service-daemons (eg. malware scanning) while a client smtp
283 connection is using TLS; with cutthrough connections this is quite likely.
285 JH/16 Fix ARC verification to do AS checks in reverse order.
287 JH/17 Support a "tls" option on the ${readsocket } expansion item.
289 JH/18 Bug 2287: Fix the protocol name (eg utf8esmtp) for multiple messages
290 using the SMTPUTF8 option on their MAIL FROM commands, in one connection.
291 Previously the "utf8" would be re-prepended for every additional message.
293 JH/19 Reject MAIL FROM commands with SMTPUTF8 when the facility was not advertised.
294 Previously thery were accepted, resulting in issues when attempting to
295 forward messages to a non-supporting MTA.
297 PP/02 Let -n work with printing macros too, not just options.
299 JH/20 Bug 2296: Fix cutthrough for >1 address redirection. Previously only
300 one parent address was copied, and bogus data was used at delivery-logging
301 time. Either a crash (after delivery) or bogus log data could result.
302 Discovery and analysis by Tim Stewart.
304 PP/03 Make ${utf8clean:} expansion operator detect incomplete final character.
305 Previously if the string ended mid-character, we did not insert the
306 promised '?' replacement.
308 PP/04 Documentation: current string operators work on bytes, not codepoints.
310 JH/21 Change as many as possible of the global flags into one-bit bitfields; these
311 should pack well giving a smaller memory footprint so better caching and
312 therefore performance. Group the declarations where this can't be done so
313 that the byte-sized flag variables are not interspersed among pointer
314 variables, giving a better chance of good packing by the compiler.
316 JH/22 Bug 1896: Fix the envelope from for DMARC forensic reports to be possibly
317 non-null, to avoid issues with sites running BATV. Previously reports were
318 sent with an empty envelope sender so looked like bounces.
320 JH/23 Bug 2318: Fix the noerror command within filters. It wasn't working.
321 The ignore_error flag wasn't being returned from the filter subprocess so
322 was not set for later routers. Investigation and fix by Matthias Kurz.
324 JH/24 Bug 2310: Raise a msg:fail:internal event for each undelivered recipient,
325 and a msg:complete for the whole, when a message is manually removed using
326 -Mrm. Developement by Matthias Kurz, hacked on by JH.
328 JH/25 Avoid fixed-size buffers for pathnames in DB access. This required using
329 a "Gnu special" function, asprintf() in the DB utility binary builds; I
330 hope that is portable enough.
332 JH/26 Bug 2311: Fix DANE-TA verification under GnuTLS. Previously it was also
333 requiring a known-CA anchor certificate; make it now rely entirely on the
334 TLSA as an anchor. Checking the name on the leaf cert against the name
335 on the A-record for the host is still done for TA (but not for EE mode).
337 JH/27 Fix logging of proxy address. Previously, a pointless "PRX=[]:0" would be
338 included in delivery lines for non-proxied connections, when compiled with
339 SUPPORT_SOCKS and running with proxy logging enabled.
341 JH/28 Bug 2314: Fire msg:fail:delivery event even when error is being ignored.
342 Developement by Matthias Kurz, tweaked by JH. While in that bit of code,
343 move the existing event to fire before the normal logging of message
344 failure so that custom logging is bracketed by normal logging.
346 JH/29 Bug 2322: A "fail" command in a non-system filter (file) now fires the
347 msg:fail:internal event. Developement by Matthias Kurz.
349 JH/30 Bug 2329: Increase buffer size used for dns lookup from 2k, which was
350 far too small for todays use of crypto signatures stored there. Go all
351 the way to the max DNS message size of 64kB, even though this might be
352 overmuch for IOT constrained device use.
354 JH/31 Fix a bad use of a copy function, which could be used to pointlessly
355 copy a string over itself. The library routine is documented as not
356 supporting overlapping copies, and on MacOS it actually raised a SIGABRT.
358 JH/32 For main options check_spool_space and check_inode_space, where the
359 platform supports 64b integers, support more than the previous 2^31 kB
360 (i.e. more than 2 TB). Accept E, P and T multipliers in addition to
361 the previous G, M, k.
363 JH/33 Bug 2338: Fix the cyrus-sasl authenticator to fill in the
364 $authenticated_fail_id variable on authentication failure. Previously
367 JH/34 Increase RSA keysize of autogen selfsign cert from 1024 to 2048. RHEL 8.0
368 OpenSSL didn't want to use such a weak key. Do for GnuTLS also, and for
369 more-modern GnuTLS move from GNUTLS_SEC_PARAM_LOW to
370 GNUTLS_SEC_PARAM_MEDIUM.
372 JH/35 OpenSSL: fail the handshake when SNI processing hits a problem, server
373 side. Previously we would continue as if no SNI had been received.
375 JH/36 Harden the handling of string-lists. When a list consisted of a sole
376 "<" character, which should be a list-separator specification, we walked
377 off past the nul-terimation.
379 JH/37 Bug 2341: Send "message delayed" warning MDNs (restricted to external
380 causes) even when the retry time is not yet met. Previously they were
381 not, meaning that when (say) an account was over-quota and temp-rejecting,
382 and multiple senders' messages were queued, only one sender would get
383 notified on each configured delay_warning cycle.
385 JH/38 Bug 2351: Log failures to extract envelope addresses from message headers.
387 JH/39 OpenSSL: clear the error stack after an SSL_accept(). With anon-auth
388 cipher-suites, an error can be left on the stack even for a succeeding
389 accept; this results in impossible error messages when a later operation
392 AM/01 Bug 2359: GnuTLS: repeat lowlevel read and write operations while they
393 return error codes indicating retry. Under TLS1.3 this becomes required.
395 JH/40 Fix the feature-cache refresh for EXPERIMENTAL_PIPE_CONNECT. Previously
396 it only wrote the new authenticators, resulting in a lack of tracking of
397 peer changes of ESMTP extensions until the next cache flush.
399 JH/41 Fix the loop reading a message header line to check for integer overflow,
400 and more-often against header_maxsize. Previously a crafted message could
401 induce a crash of the recive process; now the message is cleanly rejected.
403 JH/42 Bug 2366: Fix the behaviour of the dkim_verify_signers option. It had
404 been totally disabled for all of 4.91. Discovery and fix by "Mad Alex".
410 GF/01 DEFER rather than ERROR on redis cluster MOVED response.
411 When redis_servers is set to a list of > 1 element, and the Redis servers
412 in that list are in cluster configuration, convert the REDIS_REPLY_ERROR
413 case of MOVED into a DEFER case instead, thus moving the query onto the
414 next server in the list. For a cluster of N elements, all N servers must
415 be defined in redis_servers.
417 GF/02 Catch and remove uninitialized value warning in exiqsumm
418 Check for existence of @ARGV before looking at $ARGV[0]
420 JH/01 Replace the store_release() internal interface with store_newblock(),
421 which internalises the check required to safely use the old one, plus
422 the allocate and data copy operations duplicated in both (!) of the
423 extant use locations.
425 JH/02 Disallow '/' characters in queue names specified for the "queue=" ACL
426 modifier. This matches the restriction on the commandline.
428 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
429 Previously only the last row was returned.
431 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
432 we assumed that tags in the header were well-formed, and parsed the
433 element content after inspecting only the first char of the tag.
434 Assumptions at that stage could crash the receive process on malformed
437 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
438 While running the DKIM ACL we operate on the Permanent memory pool so that
439 variables created with "set" persist to the DATA ACL. Also (at any time)
440 DNS lookups that fail create cache records using the Permanent pool. But
441 expansions release any allocations made on the current pool - so a dnsdb
442 lookup expansion done in the DKIM ACL releases the memory used for the
443 DNS negative-cache, and bad things result. Solution is to switch to the
444 Main pool for expansions.
445 While we're in that code, add checks on the DNS cache during store_reset,
446 active in the testsuite.
447 Problem spotted, and debugging aided, by Wolfgang Breyha.
449 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
450 When none of the hosts presented to a transport match an already-open
451 connection, close it and proceed with the list. Previously we would
452 queue the message. Spotted by Lena with Yahoo, probably involving
455 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
456 Previously a spurious "250 OK id=" response was appended to the proper
459 JH/08 The "support for" informational output now, which built with Content
460 Scanning support, has a line for the malware scanner interfaces compiled
461 in. Interface can be individually included or not at build time.
463 JH/09 The "aveserver", "kavdaemon" and "mksd" interfaces are now not included
464 by the template makefile "src/EDITME". The "STREAM" support for an older
465 ClamAV interface method is removed.
467 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
468 rows affected is given instead).
470 JH/11 The runtime Berkeley DB library version is now additionally output by
471 "exim -d -bV". Previously only the compile-time version was shown.
473 JH/12 Bug 2230: Fix cutthrough routing for nonfirst messages in an initiating
474 SMTP connection. Previously, when one had more receipients than the
475 first, an abortive onward connection was made. Move to full support for
476 multiple onward connections in sequence, handling cutthrough connection
477 for all multi-message initiating connections.
479 JH/13 Bug 2229: Fix cutthrough routing for nonstandard port numbers defined by
480 routers. Previously, a multi-recipient message would fail to match the
481 onward-connection opened for the first recipient, and cause its closure.
483 JH/14 Bug 2174: A timeout on connect for a callout was also erroneously seen as
484 a timeout on read on a GnuTLS initiating connection, resulting in the
485 initiating connection being dropped. This mattered most when the callout
486 was marked defer_ok. Fix to keep the two timeout-detection methods
489 JH/15 Relax results from ACL control request to enable cutthrough, in
490 unsupported situations, from error to silently (except under debug)
491 ignoring. This covers use with PRDR, frozen messages, queue-only and
494 HS/01 Fix Buffer overflow in base64d() (CVE-2018-6789)
496 JH/16 Fix bug in DKIM verify: a buffer overflow could corrupt the malloc
497 metadata, resulting in a crash in free().
499 PP/01 Fix broken Heimdal GSSAPI authenticator integration.
500 Broken in f2ed27cf5, missing an equals sign for specified-initialisers.
501 Broken also in d185889f4, with init system revamp.
503 JH/17 Bug 2113: Fix conversation closedown with the Avast malware scanner.
504 Previously we abruptly closed the connection after reading a malware-
505 found indication; now we go on to read the "scan ok" response line,
508 JH/18 Bug 2239: Enforce non-usability of control=utf8_downconvert in the mail
509 ACL. Previously, a crash would result.
511 JH/19 Speed up macro lookups during configuration file read, by skipping non-
512 macro text after a replacement (previously it was only once per line) and
513 by skipping builtin macros when searching for an uppercase lead character.
515 JH/20 DANE support moved from Experimental to mainline. The Makefile control
516 for the build is renamed.
518 JH/21 Fix memory leak during multi-message connections using STARTTLS. A buffer
519 was allocated for every new TLS startup, meaning one per message. Fix
520 by only allocating once (OpenSSL) or freeing on TLS-close (GnuTLS).
522 JH/22 Bug 2236: When a DKIM verification result is overridden by ACL, DMARC
523 reported the original. Fix to report (as far as possible) the ACL
524 result replacing the original.
526 JH/23 Fix memory leak during multi-message connections using STARTTLS under
527 OpenSSL. Certificate information is loaded for every new TLS startup,
528 and the resources needed to be freed.
530 JH/24 Bug 2242: Fix exim_dbmbuild to permit directoryless filenames.
532 JH/25 Fix utf8_downconvert propagation through a redirect router. Previously it
535 JH/26 Bug 2253: For logging delivery lines under PRDR, append the overall
536 DATA response info to the (existing) per-recipient response info for
537 the "C=" log element. It can have useful tracking info from the
538 destination system. Patch from Simon Arlott.
540 JH/27 Bug 2251: Fix ldap lookups that return a single attribute having zero-
541 length value. Previously this would segfault.
543 HS/02 Support Avast multiline protoocol, this allows passing flags to
544 newer versions of the scanner.
546 JH/28 Ensure that variables possibly set during message acceptance are marked
547 dead before release of memory in the daemon loop. This stops complaints
548 about them when the debug_store option is enabled. Discovered specifically
549 for sender_rate_period, but applies to a whole set of variables.
550 Do the same for the queue-runner and queue-list loops, for variables set
551 from spool message files. Do the same for the SMTP per-message loop, for
552 certain variables indirectly set in ACL operations.
554 JH/29 Bug 2250: Fix a longstanding bug in heavily-pipelined SMTP input (such
555 as a multi-recipient message from a mailinglist manager). The coding had
556 an arbitrary cutoff number of characters while checking for more input;
557 enforced by writing a NUL into the buffer. This corrupted long / fast
558 input. The problem was exposed more widely when more pipelineing of SMTP
559 responses was introduced, and one Exim system was feeding another.
560 The symptom is log complaints of SMTP syntax error (NUL chars) on the
561 receiving system, and refused recipients seen by the sending system
562 (propating to people being dropped from mailing lists).
563 Discovered and pinpointed by David Carter.
565 JH/30 The (EXPERIMENTAL_DMARC) variable $dmarc_ar_header is withdrawn, being
566 replaced by the ${authresults } expansion.
568 JH/31 Bug 2257: Fix pipe transport to not use a socket-only syscall.
570 HS/03 Set a handler for SIGTERM and call exit(3) if running as PID 1. This
571 allows proper process termination in container environments.
573 JH/32 Bug 2258: Fix spool_wireformat in combination with LMTP transport.
574 Previously the "final dot" had a newline after it; ensure it is CR,LF.
576 JH/33 SPF: remove support for the "spf" ACL condition outcome values "err_temp"
577 and "err_perm", deprecated since 4.83 when the RFC-defined words
578 "temperror" and "permerror" were introduced.
580 JH/34 Re-introduce enforcement of no cutthrough delivery on transports having
581 transport-filters or DKIM-signing. The restriction was lost in the
582 consolidation of verify-callout and delivery SMTP handling.
583 Extend the restriction to also cover ARC-signing.
585 JH/35 Cutthrough: for a final-dot response timeout (and nonunderstood responses)
586 in defer=pass mode supply a 450 to the initiator. Previously the message
589 PP/02 DANE: add dane_require_tls_ciphers SMTP Transport option; if unset,
590 tls_require_ciphers is used as before.
592 HS/03 Malware Avast: Better match the Avast multiline protocol. Add
593 "pass_unscanned". Only tmpfails from the scanner are written to
594 the paniclog, as they may require admin intervention (permission
595 denied, license issues). Other scanner errors (like decompression
596 bombs) do not cause a paniclog entry.
598 JH/36 Fix reinitialisation of DKIM logging variable between messages.
599 Previously it was possible to log spurious information in receive log
602 JH/37 Bug 2255: Revert the disable of the OpenSSL session caching. This
603 triggered odd behaviour from Outlook Express clients.
605 PP/03 Add util/renew-opendmarc-tlds.sh script for safe renewal of public
608 JH/38 DKIM: accept Ed25519 pubkeys in SubjectPublicKeyInfo-wrapped form,
609 since the IETF WG has not yet settled on that versus the original
610 "bare" representation.
612 JH/39 Fix syslog logging for syslog_timestamp=no and log_selector +millisec.
613 Previously the millisecond value corrupted the output.
614 Fix also for syslog_pid=no and log_selector +pid, for which the pid
615 corrupted the output.
621 JH/01 Rework error string handling in TLS interface so that the caller in
622 more cases is responsible for logging. This permits library-sourced
623 string to be attached to addresses during delivery, and collapses
624 pairs of long lines into single ones.
626 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
627 during configuration. Wildcards are allowed and expanded.
629 JH/02 Rework error string handling in DKIM to pass more info back to callers.
630 This permits better logging.
632 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
633 do not close it down and have the child transport start it up again on
634 the passed-on TCP connection. Instead, proxy the child (and any
635 subsequent ones) for TLS via a unix-domain socket channel. Logging is
636 affected: the continued delivery log lines do not have any DNSSEC, TLS
637 Certificate or OCSP information. TLS cipher information is still logged.
639 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
640 identical IP addresses on different listening ports. Will also affect
643 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
644 add noisy ifdef guards to special-case this sillyness.
645 Patch from Bernd Kuhls.
647 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
648 than 255 are no longer allowed.
650 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
651 Disable the session-cache too, which might reduce our load. Since we
652 currrectly use a new context for every connection, both as server and
653 client, there is no benefit for these.
654 GnuTLS appears to not support tickets server-side by default (we don't
655 call gnutls_session_ticket_enable_server()) but client side is enabled
656 by default on recent versions (3.1.3 +) unless the PFS priority string
659 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
660 <https://reproducible-builds.org/specs/source-date-epoch/>.
662 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
663 the check for any unsuccessful recipients did not notice the limit, and
664 erroneously found still-pending ones.
666 JH/08 Pipeline CHUNKING command and data together, on kernels that support
667 MSG_MORE. Only in-clear (not on TLS connections).
669 JH/09 Avoid using a temporary file during transport using dkim. Unless a
670 transport-filter is involved we can buffer the headers in memory for
671 creating the signature, and read the spool data file once for the
672 signature and again for transmission.
674 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
675 4.77 as the kernel support then wasn't solid, having issues in 64bit
676 mode. Now, it's been long enough. Add support for FreeBSD also.
678 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
679 case where the routing stage had gathered several addresses to send to
680 a host before calling the transport for the first, we previously failed
681 to close down TLS in the old transport process before passing the TCP
682 connection to the new process. The new one sent a STARTTLS command
683 which naturally failed, giving a failed delivery and bloating the retry
684 database. Investigation and fix prototype from Wolfgang Breyha.
686 JH/12 Fix check on SMTP command input synchronisation. Previously there were
687 false-negatives in the check that the sender had not preempted a response
688 or prompt from Exim (running as a server), due to that code's lack of
689 awareness of the SMTP input buffering.
691 PP/04 Add commandline_checks_require_admin option.
692 Exim drops privileges sanely, various checks such as -be aren't a
693 security problem, as long as you trust local users with access to their
694 own account. When invoked by services which pass untrusted data to
695 Exim, this might be an issue. Set this option in main configuration
696 AND make fixes to the calling application, such as using `--` to stop
699 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
700 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
701 responses to those, into a single TLS record each way (this usually means
702 a single packet). As a side issue, smtp_enforce_sync now works on TLS
705 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
706 affects you only if you're dancing at the edge of the param size limits.
707 If you are, and this message makes sense to you, then: raise the
708 configured limit or use OpenSSL 1.1. Nothing we can do for older
711 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
712 cmdline element to get the documented default one. Previously it was
715 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
716 are unsuable for non-message-specific reasons - eg. connection timeout,
717 banner-time rejection.
719 JH/16 Fix logging of delivery remote port, when specified by router, under
722 PP/06 Repair manualroute's ability to take options in any order, even if one
723 is the name of a transport.
726 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
728 JH/17 Change the list-building routines interface to use the expanding-string
729 triplet model, for better allocation and copying behaviour.
731 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
732 Previously it was constructed the first time a possibly-matching string
733 was met in the configuration file input during startup; now it is done
736 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
737 compatible one, to avoid the (poorly documented) possibility of a config
738 file in the working directory redirecting the DB files, possibly correpting
739 some existing file. CVE-2017-10140 assigned for BDB.
741 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
742 cache-hot. Previously, although the result was properly cached, the
743 initial verify call returned a defer.
745 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
746 the main verify for receipient in uncached-mode.
748 JH/22 Retire historical build files to an "unsupported" subdir. These are
749 defined as "ones for which we have no current evidence of testing".
751 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
752 if present. Previously it was ignored.
754 JH/24 Start using specified-initialisers in C structure init coding. This is
755 a C99 feature (it's 2017, so now considered safe).
757 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
758 if was a fixed-sized field and bitmask ops via macros; it is now more
761 PP/07 GitHub PR 56: Apply MariaDB build fix.
762 Patch provided by Jaroslav Škarvada.
764 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
765 during Coverity cleanups [4.87 JH/47]
766 Diagnosis and fix provided by Michael Fischer v. Mollard.
768 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
769 the right size to place the terminating semicolon on its own folded
770 line, the header hash was calculated to an incorrect value thanks to
771 the (relaxed) space the fold became.
773 HS/02 Fix Bug 2130: large writes from the transport subprocess were chunked
774 and confused the parent.
776 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
777 which could crash as a result. This could lead to undeliverable messages.
779 JH/28 Logging: "next input sent too soon" now shows where input was truncated
782 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
783 matters on fast-turnover and PID-randomising systems, which were getting
784 out-of-order delivery.
786 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
787 a possibly-overlapping copy. The symptom was that "Remote host closed
788 connection in response to HELO" was logged instead of the actual 4xx
791 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
792 Previously only that bufferd was discarded, resulting in SYMTP command
795 JH/32 DKIM: when a message has multiple signatures matching an identity given
796 in dkim_verify_signers, run the dkim acl once for each. Previously only
797 one run was done. Bug 2189.
799 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
800 "panic the current process" to "deliberately defer". The panic log is
801 still written with the problem list name; the mail and reject logs now
802 get a temp-reject line for the message that was being handled, saying
803 something like "domains check lookup or other defer". The SMTP 451
804 message is still "Temporary local problem".
806 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
807 A crafted sequence of BDAT commands could result in in-use memory beeing
808 freed. CVE-2017-16943.
810 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
811 from SMTP input. Previously it was always done; now only done for DATA
812 and not BDAT commands. CVE-2017-16944.
814 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
815 to the message (such as an overlong header line). Previously this was
816 not done and we did not exit BDAT mode. Followon from the previous item
817 though a different problem.
823 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
824 than -2003 did; needs libidn2 in addition to libidn.
826 JH/02 The path option on a pipe transport is now expanded before use.
828 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
829 Patch provided by "Björn", documentation fix added too.
831 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
832 missing a wire-to-host endian conversion.
834 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
835 close after a BDAT command line could be taken as a following command,
836 giving a synch failure. Fix by only checking for synch immediately
837 before acknowledging the chunk.
839 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
840 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
841 macro. Patches provided by Josh Soref.
843 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
844 Previously we did not; the RFC seems ambiguous and VRFY is not listed
845 by IANA as a service extension. However, John Klensin suggests that we
848 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
849 the dkim code may be unix-mode line endings rather than smtp wire-format
850 CRLF, so prepend a CR to any bare LF.
852 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
853 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
855 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
856 the first were themselves being wrongly included in the feed into dkim
857 processing; with most chunk sizes in use this resulted in an incorrect
858 body hash calculated value.
860 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
861 DKIM signature block, for verification. Although advised against by
862 standards it is specifically not ruled illegal.
864 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
866 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
867 missing a body hash (the bh= tag).
869 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
870 It seems that HAProxy sends the Proxy Protocol information in clear and
871 only then does a TLS startup, so do the same.
873 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
874 TCP connections (such as for Spamd) unless the daemon successfully set
875 Fast Open mode on its listening sockets. This fixes breakage seen on
876 too-old kernels or those not configured for Fast Open, at the cost of
877 requiring both directions being enabled for TFO, and TFO never being used
878 by non-daemon-related Exim processes.
880 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
881 endings, at least on the first header line. Try to canonify any that get
882 past that check, despite the cost.
884 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
885 now limited to an arbitrary five deep, while parsing addresses with the
886 strip_excess_angle_brackets option enabled.
888 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
889 instead leave the unprompted TLS handshake in socket buffer for the
890 TLS library to consume.
892 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
894 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
896 JH/16 Drop variables when they go out of scope. Memory management drops a whole
897 region in one operation, for speed, and this leaves assigned pointers
898 dangling. Add checks run only under the testsuite which checks all
899 variables at a store-reset and panics on a dangling pointer; add code
900 explicitly nulling out all the variables discovered. Fixes one known
901 bug: a transport crash, where a dangling pointer for $sending_ip_address
902 originally assigned in a verify callout, is re-used.
904 PP/06 Drop '.' from @INC in various Perl scripts.
906 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
908 PP/08 Reduce a number of compilation warnings under clang; building with
909 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
910 should be warning-free.
912 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
914 HS/01 Fix portability problems introduced by PP/08 for platforms where
915 realloc(NULL) is not equivalent to malloc() [SunOS et al].
917 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
918 chunk. This allows us to accept broken chunked messages. We need a more
919 general solution here.
921 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
922 already-broken messages in the queue.
924 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
926 JH/19 Fix reference counting bug in routing-generated-address tracking.
932 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
933 supports it and a size is available (ie. the sending peer gave us one).
935 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
936 years of being deprecated). The replacements are the ACLs
937 acl_smtp_mime and acl_not_smtp_mime.
939 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
940 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
941 or even an in-clear connection were permitted. Now, if the host lookup
942 was dnssec and dane was requested then the host is only used if the
943 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
944 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
945 if one fails this test.
946 This means that a poorly-configured remote DNS will make it incommunicado;
947 but it protects against a DNS-interception attack on it.
949 JH/04 Bug 1810: make continued-use of an open smtp transport connection
950 non-noisy when a race steals the message being considered.
952 JH/05 If main configuration option tls_certificate is unset, generate a
953 self-signed certificate for inbound TLS connections.
955 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
956 in rewrites and routers.
958 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
959 and logged a warning sing 4.83; now they are a configuration file error.
961 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
962 (lacking @domain). Apply the same qualification processing as RCPT.
964 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
966 JH/10 Support ${sha256:} applied to a string (as well as the previous
969 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
970 a cutthrough deliver is pending, as we always want to make a connection.
971 This also avoids re-routing the message when later placing the cutthrough
972 connection after a verify cache hit.
973 Do not update it with the verify result either.
975 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
976 when routing results in more than one destination address.
978 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
979 signing (which inhibits the cutthrough capability). Previously only
980 the presence of an option was tested; now an expansion evaluating as
981 empty is permissible (obviously it should depend only on data available
982 when the cutthrough connection is made).
984 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
985 the relevant preceding SMTP command did not note the pipelining mode.
987 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
988 Previously they were not counted.
990 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
991 as one having no matching records. Previously we deferred the message
992 that needed the lookup.
994 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
995 distinguished as "(=".
997 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
998 for missing MX records. Previously it only worked for missing A records.
1000 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
1002 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
1003 after the data-go-ahead and data-ack. Patch from Jason Betts.
1005 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
1006 even for a "none" policy. Patch from Tony Meyer.
1008 JH/22 Fix continued use of a connection for further deliveries. If a port was
1009 specified by a router, it must also match for the delivery to be
1012 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
1013 When one of the recipients of a message was unsuitable for the connection
1014 (has no matching addresses), we lost track of needing to mark it
1015 deferred. As a result mail would be lost.
1017 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
1019 JH/25 Decoding ACL controls is now done using a binary search; the source code
1020 takes up less space and should be simpler to maintain. Merge the ACL
1021 condition decode tables also, with similar effect.
1023 JH/26 Fix problem with one_time used on a redirect router which returned the
1024 parent address unchanged. A retry would see the parent address marked as
1025 delivered, so not attempt the (identical) child. As a result mail would
1028 JH/27 Fix a possible security hole, wherein a process operating with the Exim
1029 UID can gain a root shell. Credit to http://www.halfdog.net/ for
1030 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
1033 JH/28 Enable {spool,log} filesystem space and inode checks as default.
1034 Main config options check_{log,spool}_{inodes,space} are now
1035 100 inodes, 10MB unless set otherwise in the configuration.
1037 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
1038 Previously it only applied to the main-section connection policy
1041 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
1043 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
1044 by me. Added RFC7919 DH primes as an alternative.
1046 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
1047 are not in the system include path.
1049 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
1050 GnuTLS, when a session startup failed (eg because the client disconnected)
1051 Exim did stdio operations after fclose. This was exposed by a recent
1052 change which nulled out the file handle after the fclose.
1054 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
1055 signed directly by the cert-signing cert, rather than an intermediate
1056 OCSP-signing cert. This is the model used by LetsEncrypt.
1058 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
1060 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
1061 an incoming connection.
1063 HS/02 Bug 1802: Do not half-close the connection after sending a request
1066 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
1067 fallback to "prime256v1".
1069 JH/34 SECURITY: Use proper copy of DATA command in error message.
1070 Could leak key material. Remotely exploitable. CVE-2016-9963.
1076 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
1077 and 3.4.4 - once the server is enabled to respond to an OCSP request
1078 it does even when not requested, resulting in a stapling non-aware
1079 client dropping the TLS connection.
1081 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
1082 support variable-length bit vectors. No functional change.
1084 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
1085 The I= interface field on outgoing lines is now after the H= remote
1086 host field, same as incoming lines. There is a separate
1087 outgoing_interface log selector which allows you to disable the
1090 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
1091 If not running log_selector +smtp_connection the mainlog would be held
1092 open indefinitely after a "too many connections" event, including to a
1093 deleted file after a log rotate. Leave the per net connection logging
1094 leaving it open for efficiency as that will be quickly detected by the
1095 check on the next write.
1097 HS/01 Bug 1671: Fix post transport crash.
1098 Processing the wait-<transport> messages could crash the delivery
1099 process if the message IDs didn't exist for some reason. When
1100 using 'split_spool_directory=yes' the construction of the spool
1101 file name failed already, exposing the same netto behaviour.
1103 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
1104 mime_regex ACL conditions.
1106 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
1107 to DSN fail messages (bounces): remote IP, remote greeting, remote response
1108 to HELO, local diagnostic string.
1110 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
1111 log line to debug. Even when configured with a tls authenticator many
1112 client connections are expected to not authenticate in this way, so
1113 an authenticate fail is not an error.
1115 HS/02 Add the Exim version string to the process info. This way exiwhat
1116 gives some more detail about the running daemon.
1118 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
1119 matter for fast-change records such as DNSBLs.
1121 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
1122 retry record, even if constant. There may be multiple transports with
1123 different interface settings and the retry behaviour needs to be kept
1126 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
1128 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
1130 JH/10 Bug 840: fix log_defer_output option of pipe transport
1132 JH/11 Bug 830: use same host for all RCPTS of a message, even under
1133 hosts_randomize. This matters a lot when combined with mua_wrapper.
1135 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
1136 ${quote_pgsql:<string>} operator.
1138 JH/13 Bug 1708: avoid misaligned access in cached lookup.
1140 JH/14 Change header file name for freeradius-client. Relevant if compiling
1141 with Radius support; from the Gentoo tree and checked under Fedora.
1143 JH/15 Bug 1712: Introduce $prdr_requested flag variable
1145 JH/16 Bug 1714: Permit an empty string as expansion result for transport
1146 option transport_filter, meaning no filtering.
1148 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
1150 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
1151 defaults to "*" (all hosts). The variable is now available when not built
1152 with TLS, default unset, mainly to enable keeping the testsuite sane.
1153 If a server certificate is not supplied (via tls_certificate) an error is
1154 logged, and clients will find TLS connections fail on startup. Presumably
1155 they will retry in-clear.
1156 Packagers of Exim are strongly encouraged to create a server certificate
1157 at installation time.
1159 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
1160 with the $config_file variable.
1162 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
1163 in transport context, after the attempt, and per-recipient. The latter type
1164 is per host attempted. The event data is the error message, and the errno
1165 information encodes the lookup type (A vs. MX) used for the (first) host,
1166 and the trailing two digits of the smtp 4xx response.
1168 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
1169 to write to mainlog (or rejectlog, paniclog) in the window between file
1170 creation and permissions/ownership being changed. Particularly affects
1171 installations where exicyclog is run as root, rather than exim user;
1172 result is that the running daemon panics and dies.
1174 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
1176 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
1177 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
1178 "pri" and "weight". Note that the previous implicit priority given by the
1179 list order is no longer honoured.
1181 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
1182 for DKIM processing.
1184 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
1185 by defining SUPPORT_SOCKS.
1187 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
1188 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
1189 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
1190 variables are renamed to proxy_{local,external}_{address,port}.
1192 JH/27 Move Internationalisation support from Experimental to mainline, enabled
1193 for a build by defining SUPPORT_I18N
1195 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
1196 of the query string, and make ${quote_redis:} do that quoting.
1198 JH/29 Move Events support from Experimental to mainline, enabled by default
1199 and removable for a build by defining DISABLE_EVENT.
1201 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
1203 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
1204 cached by the daemon.
1206 JH/32 Move Redis support from Experimental to mainline, enabled for a build
1207 by defining LOOKUP_REDIS. The libhiredis library is required.
1209 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
1210 keys are given for lookup.
1212 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
1213 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
1214 only supported when built with TLS support. The PolarSSL SHA routines
1215 are still used when the TLS library is too old for convenient support.
1217 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
1218 openssl_options), for security. OpenSSL forces this from version 1.1.0
1219 server-side so match that on older versions.
1221 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
1222 allocation for $value could be released as the expansion processing
1223 concluded, but leaving the global pointer active for it.
1225 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
1226 and to use the domains and local_parts ACL conditions.
1228 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
1229 incorrectly not doubled on cutthrough transmission, hence seen as a
1230 body-termination at the receiving system - resulting in truncated mails.
1231 Commonly the sender saw a TCP-level error, and retransmitted the message
1232 via the normal store-and-forward channel. This could result in duplicates
1233 received - but deduplicating mailstores were liable to retain only the
1234 initial truncated version.
1236 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
1238 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
1240 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
1241 we're in there, support oversigning also; bug 1309.
1243 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
1245 HS/04 Add support for keep_environment and add_environment options.
1247 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
1248 either intentional arithmetic overflow during PRNG, or testing config-
1251 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
1252 delivery resulted in actual delivery. Cancel cutthrough before DATA
1255 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
1256 rejects a recipient: pass the reject to the originator.
1258 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
1259 Many were false-positives and ignorable, but it's worth fixing the
1262 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
1263 for the new environment-manipulation done at startup. Move the routines
1264 from being local to tls.c to being global via the os.c file.
1266 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
1267 an extract embedded as result-arg for a map, the first arg for extract
1268 is unavailable so we cannot tell if this is a numbered or keyed
1269 extraction. Accept either.
1275 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
1278 JH/02 The smtp transport option "multi_domain" is now expanded.
1280 JH/03 The smtp transport now requests PRDR by default, if the server offers
1283 JH/04 Certificate name checking on server certificates, when exim is a client,
1284 is now done by default. The transport option tls_verify_cert_hostnames
1285 can be used to disable this per-host. The build option
1286 EXPERIMENTAL_CERTNAMES is withdrawn.
1288 JH/05 The value of the tls_verify_certificates smtp transport and main options
1289 default to the word "system" to access the system default CA bundle.
1290 For GnuTLS, only version 3.0.20 or later.
1292 JH/06 Verification of the server certificate for a TLS connection is now tried
1293 (but not required) by default. The verification status is now logged by
1294 default, for both outbound TLS and client-certificate supplying inbound
1297 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
1300 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
1301 Status Notification (bounce) messages are now MIME format per RFC 3464.
1302 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
1303 under the control of the dsn_advertise_hosts option, and routers may
1304 have a dsn_lasthop option.
1306 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
1307 default, modifiable by a malware= option. The list separator for
1308 the options can now be changed in the usual way. Bug 68.
1310 JH/10 The smtp_receive_timeout main option is now expanded before use.
1312 JH/11 The incoming_interface log option now also enables logging of the
1313 local interface on delivery outgoing connections.
1315 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
1316 if the interface and destination host and port all match.
1318 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
1321 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
1322 Patch from Andrew Lewis.
1324 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
1325 now supports optional time-restrictions, weighting, and priority
1326 modifiers per server. Patch originally by <rommer@active.by>.
1328 JH/16 The spamd_address main option now supports a mixed list of local
1329 and remote servers. Remote servers can be IPv6 addresses, and
1330 specify a port-range.
1332 JH/17 Bug 68: The spamd_address main option now supports an optional
1333 timeout value per server.
1335 JH/18 Bug 1581: Router and transport options headers_add/remove can
1336 now have the list separator specified.
1338 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
1341 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
1344 JH/21 Support for the A6 type of dns record is withdrawn.
1346 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
1347 rather than the verbs used.
1349 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
1350 from 255 to 1024 chars.
1352 JH/24 Verification callouts now attempt to use TLS by default.
1354 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
1355 are generic router options now. The defaults didn't change.
1357 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
1358 Original patch from Alexander Shikoff, worked over by JH.
1360 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
1361 files if ZCAT_COMMAND is not executable.
1363 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
1365 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
1367 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
1368 Normally benign, it bites when the pair was led to by a CNAME;
1369 modern usage is to not canonicalize the domain to a CNAME target
1370 (and we were inconsistent anyway for A-only vs AAAA+A).
1372 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
1374 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
1375 when evaluating $sender_host_dnssec.
1377 JH/31 Check the HELO verification lookup for DNSSEC, adding new
1378 $sender_helo_dnssec variable.
1380 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
1382 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
1384 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
1386 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
1387 documented as working, but never had. Support all but $spam_report.
1389 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
1390 added for tls authenticator.
1392 HS/03 Add perl_taintmode main config option
1398 TL/01 When running the test suite, the README says that variables such as
1399 no_msglog_check are global and can be placed anywhere in a specific
1400 test's script, however it was observed that placement needed to be near
1401 the beginning for it to behave that way. Changed the runtest perl
1402 script to read through the entire script once to detect and set these
1403 variables, reset to the beginning of the script, and then run through
1404 the script parsing/test process like normal.
1406 TL/02 The BSD's have an arc4random API. One of the functions to induce
1407 adding randomness was arc4random_stir(), but it has been removed in
1408 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
1409 function when detected.
1411 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
1412 cause callback expansion.
1414 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
1415 syntax errors in an expansion can be treated as a string instead of
1416 logging or causing an error, due to the internal use of bool_lax
1417 instead of bool when processing it.
1419 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
1420 server certificates when making smtp deliveries.
1422 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
1424 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
1426 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
1428 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
1429 Merged patch from Sebastian Wiedenroth.
1431 JH/05 Fix results-pipe from transport process. Several recipients, combined
1432 with certificate use, exposed issues where response data items split
1433 over buffer boundaries were not parsed properly. This eventually
1434 resulted in duplicates being sent. This issue only became common enough
1435 to notice due to the introduction of connection certificate information,
1436 the item size being so much larger. Found and fixed by Wolfgang Breyha.
1438 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
1439 size buffer was used, resulting in syntax errors when an expansion
1442 JH/07 Add support for directories of certificates when compiled with a GnuTLS
1443 version 3.3.6 or later.
1445 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
1446 is EXPERIMENTAL_EVENT, the main-configuration and transport options
1447 both become "event_action", the variables become $event_name, $event_data
1448 and $event_defer_errno. There is a new variable $verify_mode, usable in
1449 routers, transports and related events. The tls:cert event is now also
1450 raised for inbound connections, if the main configuration event_action
1453 TL/06 In test suite, disable OCSP for old versions of openssl which contained
1454 early OCSP support, but no stapling (appears to be less than 1.0.0).
1456 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
1457 server certificate names available under the smtp transport option
1458 "tls_verify_cert_hostname" now do not permit multi-component wildcard
1461 JH/10 Time-related extraction expansions from certificates now use the main
1462 option "timezone" setting for output formatting, and are consistent
1463 between OpenSSL and GnuTLS compilations. Bug 1541.
1465 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
1466 encoded parameter in the incoming message. Bug 1558.
1468 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
1469 include certificate info, eximon was claiming there were spoolfile
1472 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
1474 JH/14 Log delivery-related information more consistently, using the sequence
1475 "H=<name> [<ip>]" wherever possible.
1477 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
1478 are problematic for Debian distribution, omit them from the release
1481 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
1483 JH/16 Fix string representation of time values on 64bit time_t architectures.
1486 JH/17 Fix a null-indirection in certextract expansions when a nondefault
1487 output list separator was used.
1492 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
1493 checkers that were complaining about end of non-void function with no
1496 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
1497 This was a regression introduced in 4.83 by another bugfix.
1499 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1501 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1502 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1508 TF/01 Correctly close the server side of TLS when forking for delivery.
1510 When a message was received over SMTP with TLS, Exim failed to clear up
1511 the incoming connection properly after forking off the child process to
1512 deliver the message. In some situations the subsequent outgoing
1513 delivery connection happened to have the same fd number as the incoming
1514 connection previously had. Exim would try to use TLS and fail, logging
1515 a "Bad file descriptor" error.
1517 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1518 utilities have not been installed.
1520 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1521 temporary space as the ACL may create new global variables.
1523 TL/01 LDAP support uses per connection or global context settings, depending
1524 upon the detected version of the libraries at build time.
1526 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1527 to extract and use the src ip:port in logging and expansions as if it
1528 were a direct connection from the outside internet. PPv2 support was
1529 updated based on HAProxy spec change in May 2014.
1531 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1533 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1534 Properly escape header and check for NULL return.
1536 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1539 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1541 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1542 characters in header names, implemented as a verify condition.
1543 Contributed by Michael Fischer v. Mollard.
1545 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1546 results permerror and temperror. Previous values are deprecated but
1547 still accepted. In a future release, err_perm and err_temp will be
1548 completely removed, which will be a backward incompatibility if the
1549 ACL tests for either of these two old results. Patch contributed by
1550 user bes-internal on the mailing list.
1552 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1554 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1555 selectors, in both main and reject logs.
1557 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1560 JH/07 Add malware type "sock" for talking to simple daemon.
1562 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1564 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1565 routers/transports under cutthrough routing.
1567 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1568 numbers. Touch up "bool" conditional to keep the same definition.
1570 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1572 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1574 JH/12 Expand items in router/transport headers_add or headers_remove lists
1575 individually rather than the list as a whole. Bug 1452.
1577 Required for reasonable handling of multiple headers_ options when
1578 they may be empty; requires that headers_remove items with embedded
1579 colons must have them doubled (or the list-separator changed).
1581 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1582 view the policy declared in the DMARC record. Currently, $dmarc_status
1583 is a combined value of both the record presence and the result of the
1586 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1588 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1589 dnslookup router and the smtp transport (applying to the forward
1592 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1593 of ldap servers used for a specific lookup. Patch provided by Heiko
1596 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1597 New variable $lookup_dnssec_authenticated for observability.
1599 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1600 Patch submitted by Lars Timman.
1602 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1604 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1605 Requires trusted mode and valid format message id, aborts otherwise.
1606 Patch contributed by Heiko Schlichting.
1608 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1609 certextract with support for various fields. Bug 1358.
1611 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1612 is requested by default, modifiable by smtp transport option
1615 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1616 operate on certificate variables to give certificate fingerprints
1617 Also new ${sha256:cert_variable}.
1619 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1621 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1622 Christian Aistleitner.
1624 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1626 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1627 file. Patch from Wolfgang Breyha.
1629 JH/25 Expand the coverage of the delivery $host and $host_address to
1630 client authenticators run in verify callout. Bug 1476.
1632 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1633 align with daemon_smtp_ports. Bug 72.
1635 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1636 support and error reporting did not work properly.
1638 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1639 and is readable. Patch from Andrew Colin Kissa.
1641 TL/14 Enhance documentation of ${run expansion and how it parses the
1642 commandline after expansion, particularly in the case when an
1643 unquoted variable expansion results in an empty value.
1645 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1647 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1648 by renaming away from it. Addresses GH issue 15, reported by
1651 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1653 TL/15 SECURITY: prevent double expansion in math comparison functions
1654 (can expand unsanitized data). Not remotely exploitable.
1661 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1663 PP/02 Make -n do something, by making it not do something.
1664 When combined with -bP, the name of an option is not output.
1666 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1669 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1670 $sender_host_name and config options to manage this, and basic check
1673 PP/05 DSCP support for outbound connections and control modifier for inbound.
1675 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1676 (Only plugin which currently uses this is kerberos4, which nobody should
1677 be using, but we should make it available and other future plugins might
1678 conceivably use it, even though it would break NAT; stuff *should* be
1679 using channel bindings instead).
1681 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1682 name; added for Sendmail compatibility; requires admin caller.
1683 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1684 just ignore it); requires trusted caller.
1685 Also parse but ignore: -Ac -Am -X<logfile>
1688 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1690 TL/02 Add +smtp_confirmation as a default logging option.
1692 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1693 Patch by Magnus Holmgren from 2007-02-20.
1695 TL/04 Bugzilla 1281 - Spec typo.
1696 Bugzilla 1283 - Spec typo.
1697 Bugzilla 1290 - Spec grammar fixes.
1699 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1701 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1703 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1704 mailing list by by Dmitry Isaikin.
1706 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1708 JH/02 Support "G" suffix to numbers in ${if comparisons.
1710 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1712 NM/01 Bugzilla 1197 - Spec typo
1713 Bugzilla 1196 - Spec examples corrections
1715 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1717 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1718 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1721 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1722 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1724 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1725 "acl {{name}{arg}...}", and optional args on acl condition
1728 JH/05 Permit multiple router/transport headers_add/remove lines.
1730 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1732 JH/07 Avoid using a waiting database for a single-message-only transport.
1733 Performance patch from Paul Fisher. Bugzilla 1262.
1735 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1738 JH/09 Add $headers_added variable, with content from use of ACL modifier
1739 add_header (but not yet added to the message). Bugzilla 199.
1741 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1742 Pulled from Bugzilla 817 by Wolfgang Breyha.
1744 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1746 (nb: this is the same fix as in Exim 4.80.1)
1748 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1751 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1754 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1756 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1757 advertise SMTP AUTH mechanism to us, instead of a generic
1758 protocol violation error. Also, make Exim more robust to bad
1759 data from the Dovecot auth socket.
1761 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1763 When a queue runner is handling a message, Exim first routes the
1764 recipient addresses, during which it prunes them based on the retry
1765 hints database. After that it attempts to deliver the message to
1766 any remaining recipients. It then updates the hints database using
1769 So if a recipient address works intermittently, it can get repeatedly
1770 deferred at routing time. The retry hints record remains fresh so the
1771 address never reaches the final cutoff time.
1773 This is a fairly common occurrence when a user is bumping up against
1774 their storage quota. Exim had some logic in its local delivery code
1775 to deal with this. However it did not apply to per-recipient defers
1776 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1778 This change adds a proper retry rule check during routing so that the
1779 final cutoff time is checked against the message's age. We only do
1780 this check if there is an address retry record and there is not a
1781 domain retry record; this implies that previous attempts to handle
1782 the address had the retry_use_local_parts option turned on. We use
1783 this as an approximation for the destination being like a local
1784 delivery, as in LMTP.
1786 I suspect this new check makes the old local delivery cutoff check
1787 redundant, but I have not verified this so I left the code in place.
1789 TF/02 Correct gecos expansion when From: is a prefix of the username.
1791 Test 0254 submits a message to Exim with the header
1795 When I ran the test suite under the user fanf2, Exim expanded
1796 the header to contain my full name, whereas it should have added
1797 a Resent-Sender: header. It erroneously treats any prefix of the
1798 username as equal to the username.
1800 This change corrects that bug.
1802 GF/01 DCC debug and logging tidyup
1803 Error conditions log to paniclog rather than rejectlog.
1804 Debug lines prefixed by "DCC: " to remove any ambiguity.
1806 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1808 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1809 Bug spotted by Jeremy Harris; was flawed since initial commit.
1810 Would have resulted in OCSP responses post-SNI triggering an Exim
1811 NULL dereference and crash.
1813 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1815 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1816 Bug detection, analysis and fix by Samuel Thibault.
1817 Bugzilla 1331, Debian bug #698092.
1819 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1821 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1822 Server implementation by Todd Lyons, client by JH.
1823 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1824 config variable "prdr_enable" controls whether the server
1825 advertises the facility. If the client requests PRDR a new
1826 acl_data_smtp_prdr ACL is called once for each recipient, after
1827 the body content is received and before the acl_smtp_data ACL.
1828 The client is controlled by both of: a hosts_try_prdr option
1829 on the smtp transport, and the server advertisement.
1830 Default client logging of deliveries and rejections involving
1831 PRDR are flagged with the string "PRDR".
1833 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1834 fclose(). Diagnosis by Todd Lyons.
1836 PP/17 Update configure.default to handle IPv6 localhost better.
1837 Patch by Alain Williams (plus minor tweaks).
1840 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1841 This is now consistent with GnuTLS, and is now documented: the
1842 previous undocumented portable approach to treating the option as
1843 unset was to force an expansion failure. That still works, and
1844 an empty string is now equivalent.
1846 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1847 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1848 not performing validation itself.
1850 PP/20 Added force_command boolean option to pipe transport.
1851 Patch from Nick Koston, of cPanel Inc.
1853 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1856 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1858 PP/21 Fix eximon continuous updating with timestamped log-files.
1859 Broken in a format-string cleanup in 4.80, missed when I repaired the
1860 other false fix of the same issue.
1861 Report and fix from Heiko Schlichting.
1864 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1865 Report from Prashanth Katuri.
1867 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1868 It's SecureTransport, so affects any MacOS clients which use the
1869 system-integrated TLS libraries, including email clients.
1871 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1872 using a MIME ACL for non-SMTP local injection.
1873 Report and assistance in diagnosis by Warren Baker.
1875 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1877 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1879 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1880 last id that failed so it may be referenced in subsequent ACL's.
1882 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1885 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1886 ldap library initialization, allowing self-signed CA's to be
1887 used. Also properly sets require_cert option later in code by
1888 using NULL (global ldap config) instead of ldap handle (per
1889 session). Bug diagnosis and testing by alxgomz.
1891 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1892 the src/util/ subdirectory.
1894 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1895 renamed to Transport Post Delivery Action by Jeremy Harris, as
1898 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1899 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1900 redis_servers = needs to be configured which will be used by the redis
1901 lookup. Patch from Warren Baker, of The Packet Hub.
1903 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1905 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1906 hostname or reverse DNS when processing a host list. Used suggestions
1907 from multiple comments on this bug.
1909 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1911 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1912 Missed a few lines, added it to make the runtest require no keyboard
1915 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1916 contains upper case chars. Make router use caseful_local_part.
1918 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1919 support when GnuTLS has been built with p11-kit.
1925 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1927 This, or similar/improved, will also be change PP/11 of 4.82.
1933 PP/01 Handle short writes when writing local log-files.
1934 In practice, only affects FreeBSD (8 onwards).
1935 Bugzilla 1053, with thanks to Dmitry Isaikin.
1937 NM/01 Bugzilla 949 - Documentation tweak
1939 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1942 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1944 PP/02 Implemented gsasl authenticator.
1946 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1948 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1949 `pkg-config foo` for cflags/libs.
1951 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1952 with rest of GSASL and with heimdal_gssapi.
1954 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1955 `pkg-config foo` for cflags/libs for the TLS implementation.
1957 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1958 properties get this fed in as external SSF. A number of robustness
1959 and debugging improvements to the cyrus_sasl authenticator.
1961 PP/08 cyrus_sasl server now expands the server_realm option.
1963 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1964 Patch by Jeremy Harris.
1966 PP/10 Added dbmjz lookup type.
1968 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1970 PP/12 MAIL args handles TAB as well as SP, for better interop with
1971 non-compliant senders.
1972 Analysis and variant patch by Todd Lyons.
1974 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1975 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1976 Patch from Dirk Mueller <dmueller@suse.com>
1978 PP/13 tls_peerdn now print-escaped for spool files.
1979 Observed some $tls_peerdn in wild which contained \n, which resulted
1980 in spool file corruption.
1982 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1983 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1984 or write after TLS renegotiation, which otherwise led to messages
1987 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1988 as a tracking header (ie: a signed header comes before the signature).
1989 Patch from Wolfgang Breyha.
1991 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1992 comma-sep list; embedded commas doubled.
1994 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1996 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1998 Report and patch from Dmitry Banschikov.
2000 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
2001 Removed SSL_clear() after SSL_new() which led to protocol negotiation
2002 failures. We appear to now support TLS1.1+ with Exim.
2004 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
2005 lets Exim select keys and certificates based upon TLS SNI from client.
2006 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
2007 before an outbound SMTP session. New log_selector, +tls_sni.
2009 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
2010 NULL dereference. Report and patch from Alun Jones.
2012 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
2013 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
2014 Not seeing resolver debug output on NetBSD, but suspect this is a
2015 resolver implementation change.
2017 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
2018 Left warnings. Added "eximon gdb" invocation mode.
2020 PP/21 Defaulting "accept_8bitmime" to true, not false.
2022 PP/22 Added -bw for inetd wait mode support.
2024 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
2025 locate the relevant includes and libraries. Made this the default.
2027 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
2028 Bugzilla 1246, report and most of solution from Tomasz Kusy.
2030 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
2031 This may cause build issues on older platforms.
2033 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
2034 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
2035 gnutls_require_mac & gnutls_require_protocols (no longer supported).
2036 Added SNI support via GnuTLS too.
2037 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
2039 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
2041 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
2042 Applied second patch from Janne, implementing suggestion to default
2043 multiple-strings-in-record handling to match SPF spec.
2045 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
2047 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
2048 read-only, out of scope).
2049 Patch from Wolfgang Breyha, report from Stuart Northfield.
2051 PP/29 Fix three issues highlighted by clang analyser static analysis.
2052 Only crash-plausible issue would require the Cambridge-specific
2053 iplookup router and a misconfiguration.
2054 Report from Marcin Mirosław.
2056 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
2058 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
2059 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
2060 As part of this, removing so much warning spew let me fix some minor
2061 real issues in debug logging.
2063 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
2064 assignment on my part. Fixed.
2066 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
2067 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
2068 Janne Snabb (who went above and beyond: thank you).
2070 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
2071 string otherwise requires a connection and a bunch more work and it's
2072 relatively easy to get wrong. Should also expose TLS library linkage
2075 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
2076 64-bit ${eval} (JH/03).
2078 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
2079 GNU libc to support some of the 64-bit stuff, should not lead to
2080 conflicts. Defined before os.h is pulled in, so if a given platform
2081 needs to override this, it can.
2083 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
2084 protection layer was required, which is not implemented.
2085 Bugzilla 1254, patch from Wolfgang Breyha.
2087 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
2088 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
2089 tls_dhparam take prime identifiers. Also unbreak combination of
2090 OpenSSL+DH_params+TLSSNI.
2092 PP/39 Disable SSLv2 by default in OpenSSL support.
2098 PP/01 Solaris build fix for Oracle's LDAP libraries.
2099 Bugzilla 1109, patch from Stephen Usher.
2101 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
2103 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
2106 TF/02 Fix a couple more cases where we did not log the error message
2107 when unlink() failed. See also change 4.74-TF/03.
2109 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
2110 lock up or crash if it happened to be inside a call to libc when it
2111 got a SIGUSR1 from exiwhat.
2113 The SIGUSR1 handler appends the current process status to the process
2114 log which is later printed by exiwhat. It used to use the general
2115 purpose logging code to do this, but several functions it calls are
2116 not safe for signals.
2118 The new output code in the SIGUSR1 handler is specific to the process
2119 log, and simple enough that it's easy to inspect for signal safety.
2120 Removing some special cases also simplifies the general logging code.
2121 Removing the spurious timestamps from the process log simplifies
2124 TF/04 Improved ratelimit ACL condition.
2126 The /noupdate option has been deprecated in favour of /readonly which
2127 has clearer semantics. The /leaky, /strict, and /readonly update modes
2128 are mutually exclusive. The update mode is no longer included in the
2129 database key; it just determines when the database is updated. (This
2130 means that when you upgrade Exim will forget old rate measurements.)
2132 Exim now checks that the per_* options are used with an update mode that
2133 makes sense for the current ACL. For example, when Exim is processing a
2134 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
2135 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
2136 must specify per_mail/readonly. If you omit the update mode it defaults to
2137 /leaky where that makes sense (as before) or /readonly where required.
2139 The /noupdate option is now undocumented but still supported for
2140 backwards compatibility. It is equivalent to /readonly except that in
2141 ACLs where /readonly is required you may specify /leaky/noupdate or
2142 /strict/noupdate which are treated the same as /readonly.
2144 A useful new feature is the /count= option. This is a generalization
2145 of the per_byte option, so that you can measure the throughput of other
2146 aggregate values. For example, the per_byte option is now equivalent
2147 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
2149 The per_rcpt option has been generalized using the /count= mechanism
2150 (though it's more complicated than the per_byte equivalence). When it is
2151 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
2152 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
2153 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
2154 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
2155 Note that using per_rcpt with a non-readonly update mode in more than
2156 one ACL will cause the recipients to be double-counted. (The per_mail
2157 and per_byte options don't have this problem.)
2159 The handling of very low rates has changed slightly. If the computed rate
2160 is less than the event's count (usually one) then this event is the first
2161 after a long gap. In this case the rate is set to the same as this event's
2162 count, so that the first message of a spam run is counted properly.
2164 The major new feature is a mechanism for counting the rate of unique
2165 events. The new per_addr option counts the number of different
2166 recipients that someone has sent messages to in the last time period. It
2167 behaves like per_rcpt if all the recipient addresses are different, but
2168 duplicate recipient addresses do not increase the measured rate. Like
2169 the /count= option this is a general mechanism, so the per_addr option
2170 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
2171 example, measure the rate that a client uses different sender addresses
2172 with the options per_mail/unique=$sender_address. There are further
2173 details in the main documentation.
2175 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
2177 TF/06 Removed a few PCRE remnants.
2179 TF/07 Automatically extract Exim's version number from tags in the git
2180 repository when doing development or release builds.
2182 PP/02 Raise smtp_cmd_buffer_size to 16kB.
2183 Bugzilla 879. Patch from Paul Fisher.
2185 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
2186 Heavily based on revision 40f9a89a from Simon Arlott's tree.
2189 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
2191 PP/05 Variable $av_failed, true if the AV scanner deferred.
2192 Bugzilla 1078. Patch from John Horne.
2194 PP/06 Stop make process more reliably on build failure.
2195 Bugzilla 1087. Patch from Heiko Schlittermann.
2197 PP/07 Make maildir_use_size_file an _expandable_ boolean.
2198 Bugzilla 1089. Patch from Heiko Schlittermann.
2200 PP/08 Handle ${run} returning more data than OS pipe buffer size.
2201 Bugzilla 1131. Patch from Holger Weiß.
2203 PP/09 Handle IPv6 addresses with SPF.
2204 Bugzilla 860. Patch from Wolfgang Breyha.
2206 PP/10 GnuTLS: support TLS 1.2 & 1.1.
2208 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
2211 PP/11 match_* no longer expand right-hand-side by default.
2212 New compile-time build option, EXPAND_LISTMATCH_RHS.
2213 New expansion conditions, "inlist", "inlisti".
2215 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
2217 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
2219 PP/14 fix log_write() format string regression from TF/03.
2220 Bugzilla 1152. Patch from Dmitry Isaikin.
2226 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
2228 PP/02 Harmonised TLS library version reporting; only show if debugging.
2229 Layout now matches that introduced for other libraries in 4.74 PP/03.
2231 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
2233 PP/04 New "dns_use_edns0" global option.
2235 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
2238 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
2239 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
2241 TK/01 Updated PolarSSL code to 0.14.2.
2242 Bugzilla 1097. Patch from Andreas Metzler.
2244 PP/07 Catch divide-by-zero in ${eval:...}.
2245 Fixes bugzilla 1102.
2247 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
2250 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
2251 format-string attack -- SECURITY: remote arbitrary code execution.
2253 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
2254 time unintentionally subject to list matching rules, letting the header
2255 cause arbitrary Exim lookups (of items which can occur in lists, *not*
2256 arbitrary string expansion). This allowed for information disclosure.
2258 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
2259 INT_MIN/-1 -- value coerced to INT_MAX.
2265 NM/01 Workaround for PCRE version dependency in version reporting
2268 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
2269 This fixes portability to compilers other than gcc, notably
2270 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
2272 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
2273 makefiles for portability to HP-UX and POSIX correctness.
2275 PP/01 Permit LOOKUP_foo enabling on the make command-line.
2276 Also via indented variable definition in the Makefile.
2277 (Debugging by Oliver Heesakkers).
2279 PP/02 Restore caching of spamd results with expanded spamd_address.
2280 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
2282 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
2283 Improves build reliability. Fix from: Frank Elsner
2285 NM/02 Fix wide character breakage in the rfc2047 coding
2286 Fixes bug 1064. Patch from Andrey N. Oktyabrski
2288 NM/03 Allow underscore in dnslist lookups
2289 Fixes bug 1026. Patch from Graeme Fowler
2291 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
2292 Code patches from Adam Ciarcinski of NetBSD.
2294 NM/04 Fixed exiqgrep to cope with mailq missing size issue
2297 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
2298 is logged, to avoid truncation. Patch from John Horne.
2300 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
2301 Patch from Jakob Hirsch.
2303 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
2304 SQL string expansion failure details.
2305 Patch from Andrey Oktyabrski.
2307 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
2308 Patch from Simon Arlott.
2310 PP/09 New lookups functionality failed to compile on old gcc which rejects
2311 extern declarations in function scope.
2312 Patch from Oliver Fleischmann
2314 PP/10 Use sig_atomic_t for flags set from signal handlers.
2315 Check getgroups() return and improve debugging.
2316 Fixed developed for diagnosis in bug 927 (which turned out to be
2319 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
2320 Patch from Mark Zealey.
2322 PP/12 Bugzilla 1056: Improved spamd server selection.
2323 Patch from Mark Zealey.
2325 PP/13 Bugzilla 1086: Deal with maildir quota file races.
2326 Based on patch from Heiko Schlittermann.
2328 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
2329 Patch from Uwe Doering, sign-off by Michael Haardt.
2331 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
2332 variable declaration deep within a block. Bug and patch from
2335 PP/15 lookups-Makefile IRIX compatibility coercion.
2337 PP/16 Make DISABLE_DKIM build knob functional.
2339 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
2340 Patch by Simon Arlott
2342 TF/03 Fix valgrind.h portability to C89 compilers that do not support
2343 variable argument macros. Our copy now differs from upstream.
2349 TF/01 Failure to get a lock on a hints database can have serious
2350 consequences so log it to the panic log.
2352 TF/02 Log LMTP confirmation messages in the same way as SMTP,
2353 controlled using the smtp_confirmation log selector.
2355 TF/03 Include the error message when we fail to unlink a spool file.
2357 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
2358 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
2359 for maintaining out-of-tree patches for some time.
2361 PP/01 Bugzilla 139: Documentation and portability issues.
2362 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
2363 Handle per-OS dynamic-module compilation flags.
2365 PP/02 Let /dev/null have normal permissions.
2366 The 4.73 fixes were a little too stringent and complained about the
2367 permissions on /dev/null. Exempt it from some checks.
2368 Reported by Andreas M. Kirchwitz.
2370 PP/03 Report version information for many libraries, including
2371 Exim version information for dynamically loaded libraries. Created
2372 version.h, now support a version extension string for distributors
2373 who patch heavily. Dynamic module ABI change.
2375 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
2376 privilege escalation vulnerability whereby the Exim run-time user
2377 can cause root to append content of the attacker's choosing to
2380 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
2383 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
2384 If dropping privileges for untrusted macros, we disabled normal logging
2385 on the basis that it would fail; for the Exim run-time user, this is not
2386 the case, and it resulted in successful deliveries going unlogged.
2387 Fixed. Reported by Andreas Metzler.
2393 PP/01 Date: & Message-Id: revert to normally being appended to a message,
2394 only prepend for the Resent-* case. Fixes regression introduced in
2395 Exim 4.70 by NM/22 for Bugzilla 607.
2397 PP/02 Include check_rfc2047_length in configure.default because we're seeing
2398 increasing numbers of administrators be bitten by this.
2400 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
2402 PP/03 Bugzilla 994: added openssl_options main configuration option.
2404 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
2406 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
2408 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
2410 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
2411 without a peer certificate, leading to a segfault because of an
2412 assumption that peers always have certificates. Be a little more
2413 paranoid. Problem reported by Martin Tscholak.
2415 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
2416 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
2417 NB: ClamAV planning to remove STREAM in "middle of 2010".
2418 CL also introduces -bmalware, various -d+acl logging additions and
2419 more caution in buffer sizes.
2421 PP/09 Implemented reverse_ip expansion operator.
2423 PP/10 Bugzilla 937: provide a "debug" ACL control.
2425 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
2427 PP/12 Bugzilla 973: Implement --version.
2429 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
2431 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
2433 PP/15 Bugzilla 816: support multiple condition rules on Routers.
2435 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
2436 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
2437 ignore trailing whitespace.
2439 JJ/02 prevent non-panic DKIM error from being sent to paniclog
2441 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
2444 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
2445 Notification from Dr Andrew Aitchison.
2447 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
2448 ExtendedDetectionInfo response format.
2449 Notification from John Horne.
2451 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
2454 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
2455 XSL and documented dependency on system catalogs, with examples of how
2458 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
2461 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
2462 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
2463 configuration file which is writeable by the Exim user or group.
2465 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
2466 of configuration files to cover files specified with the -C option if
2467 they are going to be used with root privileges, not just the default
2470 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
2471 option (effectively making it always true).
2473 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
2474 files to be used while preserving root privileges.
2476 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
2477 that rogue child processes cannot use them.
2479 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
2480 run-time user, instead of root.
2482 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
2483 Exim run-time user without dropping privileges.
2485 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
2486 result string, instead of calling string_vformat() twice with the same
2489 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
2490 for other users. Others should always drop root privileges if they use
2491 -C on the command line, even for a whitelisted configure file.
2493 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
2495 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2501 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2502 $data_path, and $header_path variables; fixed documentation bugs and
2505 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2506 exipick to access non-standard spools, including the "frozen" queue
2509 NM/01 Bugzilla 965: Support mysql stored procedures.
2510 Patch from Alain Williams
2512 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2514 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2515 Patch from Andreas Metzler
2517 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2518 Patch from Kirill Miazine
2520 NM/05 Bugzilla 671: Added umask to procmail example.
2522 JJ/03 installed exipick 20100323.0, fixing doc bug
2524 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2525 directory. Notification and patch from Dan Rosenberg.
2527 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2529 TK/02 Improve log output when DKIM signing operation fails.
2531 MH/01 Treat the transport option dkim_domain as a colon separated
2532 list, not as a single string, and sign the message with each element,
2533 omitting multiple occurences of the same signer.
2535 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2536 Bugzilla 985, 986. Patch by Simon Arlott
2538 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2539 Patch by Simon Arlott
2541 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2542 MBX locking. Notification from Dan Rosenberg.
2548 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2550 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2552 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2554 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2556 NM/04 Bugzilla 915: Flag broken perl installation during build.
2562 TK/01 Added patch by Johannes Berg that expands the main option
2563 "spamd_address" if it starts with a dollar sign.
2565 TK/02 Write list of recipients to X-Envelope-Sender header when building
2566 the mbox-format spool file for content scanning (suggested by Jakob
2569 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2570 (http://www.dcc-servers.net/) support via dccifd. Activated by
2571 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2573 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2574 by Mark Daniel Reidel <mr@df.eu>.
2576 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2577 When building exim an external PCRE library is now needed -
2578 PCRE is a system library on the majority of modern systems.
2579 See entry on PCRE_LIBS in EDITME file.
2581 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2582 conversation. Added nologin parameter to request.
2583 Patch contributed by Kirill Miazine.
2585 TF/01 Do not log submission mode rewrites if they do not change the address.
2587 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2589 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2590 log files in place. Contributed by Roberto Lima.
2592 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2594 TF/03 Bugzilla 615: When checking the local_parts router precondition
2595 after a local_part_suffix or local_part_prefix option, Exim now
2596 does not use the address's named list lookup cache, since this
2597 contains cached lookups for the whole local part.
2599 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2600 Robert Millan. Documentation is in experimental-spec.txt.
2602 TF/04 Bugzilla 668: Fix parallel build (make -j).
2604 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2606 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2607 Patch provided by Jan Srzednicki.
2609 TF/05 Leading white space used to be stripped from $spam_report which
2610 wrecked the formatting. Now it is preserved.
2612 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2613 that they are available at delivery time.
2615 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2617 TF/08 TLS error reporting now respects the incoming_interface and
2618 incoming_port log selectors.
2620 TF/09 Produce a more useful error message if an SMTP transport's hosts
2621 setting expands to an empty string.
2623 NM/06 Bugzilla 744: EXPN did not work under TLS.
2624 Patch provided by Phil Pennock.
2626 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2627 Patch provided by Richard Godbee.
2629 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2630 acl_smtp_notquit, added index entry.
2632 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2633 Patch provided by Eugene Bujak.
2635 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2636 accept(). Patch provided by Maxim Dounin.
2638 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2639 Patch provided by Phil Pennock.
2641 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2643 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2644 Patch provided by Brad "anomie" Jorsch.
2646 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2647 Patch provided by Dean Brooks.
2649 TK/05 Add native DKIM support (does not depend on external libraries).
2651 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2652 Patch provided by Graeme Fowler.
2654 NM/16 Bugzilla 851: Documentation example syntax fix.
2656 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2658 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2661 NM/19 Bugzilla 745: TLS version reporting.
2662 Patch provided by Phil Pennock.
2664 NM/20 Bugzilla 167: bool: condition support.
2665 Patch provided by Phil Pennock.
2667 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2668 clients. Patch provided by Phil Pennock.
2670 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2671 Patch provided by Brad "anomie" Jorsch.
2673 NM/23 Bugzilla 687: Fix misparses in eximstats.
2674 Patch provided by Heiko Schlittermann.
2676 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2677 Patch provided by Heiko Schlittermann.
2679 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2680 plus update to original patch.
2682 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2684 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2685 Patch provided by David Brownlee.
2687 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2689 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2691 NM/30 Bugzilla 888: TLS documentation bugfixes.
2693 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2695 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2696 Unlike the original bugzilla I have changed all shell scripts in src tree.
2698 NM/33 Bugzilla 898: Transport filter timeout fix.
2699 Patch by Todd Rinaldo.
2701 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2702 Patch by Serge Demonchaux.
2704 NM/35 Bugzilla 39: Base64 decode bug fixes.
2705 Patch by Jakob Hirsch.
2707 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2709 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2711 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2713 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2719 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2720 ALT-N's libdkim that I have put here:
2721 http://duncanthrax.net/exim-experimental/
2723 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2724 were called 'true' and it seems that C99 defines that as a reserved
2725 keyword to be used with 'bool' variable types. That means you could
2726 not include C99-style headers which use bools without triggering
2727 build errors in sieve.c.
2729 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2730 as mailq or other aliases. Changed the --help handling significantly
2731 to do whats expected. exim_usage() emits usage/help information.
2733 SC/01 Added the -bylocaldomain option to eximstats.
2735 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2737 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2739 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2745 PH/01 Another patch from the Sieve maintainer.
2747 PH/02 When an IPv6 address is converted to a string for single-key lookup
2748 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2749 dots are used instead of colons so that keys in lsearch files need not
2750 contain colons. This was done some time before quoting was made available
2751 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2752 (notated using the quote facility) so as to distinguish them from IPv4
2753 keys. This meant that lookups for IP addresses in host lists did not work
2754 for iplsearch lookups.
2756 This has been fixed by arranging for IPv6 addresses to be expressed with
2757 colons if the lookup type is iplsearch. This is not incompatible, because
2758 previously such lookups could never work.
2760 The situation is now rather anomalous, since one *can* have colons in
2761 ordinary lsearch keys. However, making the change in all cases is
2762 incompatible and would probably break a number of configurations.
2764 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2767 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2768 conversion specification without a maximum field width, thereby enabling
2769 a rogue spamd server to cause a buffer overflow. While nobody in their
2770 right mind would setup Exim to query an untrusted spamd server, an
2771 attacker that gains access to a server running spamd could potentially
2772 exploit this vulnerability to run arbitrary code as the Exim user.
2774 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2775 $primary_hostname instead of what libspf2 thinks the hosts name is.
2777 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2778 a directory entry by the name of the lookup key. Previously, if a
2779 symlink pointed to a non-existing file or a file in a directory that
2780 Exim lacked permissions to read, a lookup for a key matching that
2781 symlink would fail. Now it is enough that a matching directory entry
2782 exists, symlink or not. (Bugzilla 503.)
2784 PH/03 The body_linecount and body_zerocount variables are now exported in the
2787 PH/04 Added the $dnslist_matched variable.
2789 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2790 This means they are set thereafter only if the connection becomes
2793 PH/06 Added the client_condition to authenticators so that some can be skipped
2794 by clients under certain conditions.
2796 PH/07 The error message for a badly-placed control=no_multiline_responses left
2797 "_responses" off the end of the name.
2799 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2801 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2802 (without spaces) instead of just copying the configuration text.
2804 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2806 PH/11 Added $max_received_linelength.
2808 PH/12 Added +ignore_defer and +include_defer to host lists.
2810 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2811 way in which PCRE > 7.0 is built.
2813 PH/14 Implemented queue_only_load_latch.
2815 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2816 MAIL command. The effect was to mangle the value on 64-bit systems.
2818 PH/16 Another patch from the Sieve maintainer.
2820 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2822 PH/18 If a system quota error occurred while trying to create the file for
2823 a maildir delivery, the message "Mailbox is full" was not appended to the
2824 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2825 only to a quota excession during the actual writing of the file.
2827 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2828 characters?) which causes problems in log lines. The DN values are now
2829 passed through string_printing() before being added to log lines.
2831 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2832 and InterBase are left for another time.)
2834 PH/21 Added message_body_newlines option.
2836 PH/22 Guard against possible overflow in moan_check_errorcopy().
2838 PH/23 POSIX allows open() to be a macro; guard against that.
2840 PH/24 If the recipient of an error message contained an @ in the local part
2841 (suitably quoted, of course), incorrect values were put in $domain and
2842 $local_part during the evaluation of errors_copy.
2848 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2849 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2852 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2853 issue a MAIL command.
2855 PH/02 In an ACL statement such as
2857 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2859 if a client was not listed at all, or was listed with a value other than
2860 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2861 the condition was not true (as it should be), so access was not denied.
2862 The bug was that the ! inversion was incorrectly passed on to the second
2863 item. This has been fixed.
2865 PH/03 Added additional dnslists conditions == and =& which are different from
2866 = and & when the dns lookup returns more than one IP address.
2868 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2869 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2871 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2872 FSYNC, which compiles an option called disable_fsync that allows for
2873 bypassing fsync(). The documentation is heavily laced with warnings.
2875 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2877 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2878 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2879 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2880 including adding "make clean"; (3) Added -fPIC when compiling the test
2881 dynamically loaded module, to get rid of a warning.
2883 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2884 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2885 = 0s. The bug is otherwise harmless.
2887 PH/07 There was a bug in the dovecot authenticator such that the value of
2888 $auth1 could be overwritten, and so not correctly preserved, after a
2889 successful authentication. This usually meant that the value preserved by
2890 the server_setid option was incorrect.
2892 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2894 PH/09 Installed PCRE release 7.0.
2896 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2897 run for batched SMTP input. It is now run at the start of every message
2898 in the batch. While fixing this I discovered that the process information
2899 (output by running exiwhat) was not always getting set for -bs and -bS
2900 input. This is fixed, and it now also says "batched" for BSMTP.
2902 PH/11 Added control=no_pipelining.
2904 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2905 patch, slightly modified), and move the expansion of helo_data till after
2906 the connection is made in the smtp transport (so it can use these
2909 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2911 PH/14 Added log_selector = +pid.
2913 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2915 PH/16 Add ${if forany and ${if forall.
2917 PH/17 Added dsn_from option to vary the From: line in DSNs.
2919 PH/18 Flush SMTP output before performing a callout, unless control =
2920 no_callout_flush is set.
2922 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2923 was true (the default) a successful delivery failed to delete the retry
2924 item, thus causing premature timeout of the address. The bug is now
2927 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2929 PH/21 Long custom messages for fakedefer and fakereject are now split up
2930 into multiline responses in the same way that messages for "deny" and
2931 other ACL rejections are.
2933 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2934 with slight modification.
2936 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2937 draft, changing the syntax and factoring some duplicate code".
2939 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2940 for deliveries of the second and subsequent messages over the same SMTP
2943 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2944 ${reduce, with only minor "tidies".
2946 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2948 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2949 expansion side effects.
2951 PH/27 When a message times out after an over-quota error from an Exim-imposed
2952 quota, the bounce message says "mailbox is full". This message was not
2953 being given when it was a system quota that was exceeded. It now should
2956 MH/03 Made $recipients available in local_scan(). local_scan() already has
2957 better access to the recipient list through recipients_list[], but
2958 $recipients can be useful in postmaster-provided expansion strings.
2960 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2961 in the case of a MAIL command with additional options following the
2962 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2963 were accidentally chopped off.
2965 PH/29 SMTP synchronization checks are implemented when a command is read -
2966 there is a check that no more input is waiting when there shouldn't be
2967 any. However, for some commands, a delay in an ACL can mean that it is
2968 some time before the response is written. In this time, more input might
2969 arrive, invalidly. So now there are extra checks after an ACL has run for
2970 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2971 pipelining has not been advertised.
2973 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2975 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2976 returned in the SMTP response when smtp_return_error_details was false.
2977 This has been fixed.
2979 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2980 instead of the C I/O that was originally supplied, because problems were
2981 reported on Solaris.
2983 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2984 Exim which did not show up earlier: it was assuming that a call to
2985 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2986 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2987 was a macro that became an assignment, so it seemed to work. This has
2988 changed to a proper function call with a void return, hence the compile
2989 error. Exim's code has been fixed.
2991 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2994 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2996 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2998 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2999 because it turns out to be too big or there is a timeout) there is no
3000 "Completed" line in the log. When some messages of this type were
3001 selected by exigrep, they were listed as "not completed". Others were
3002 picked up by some special patterns. I have improved the selection
3003 criteria to be more general.
3005 PH/38 The host_find_failed option in the manualroute router can now be set
3006 to "ignore", to completely ignore a host whose IP address cannot be
3007 found. If all hosts are ignored, the behaviour is controlled by the new
3008 host_all_ignored option.
3010 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
3011 homing or because of multiple MX records with /mx) generated more than
3012 one IP address, and the following item turned out to be the local host,
3013 all the secondary addresses of the first item were incorrectly removed
3014 from the list, along with the local host and any following hosts (which
3015 is what is supposed to happen).
3017 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
3018 whoever called Exim into the -H file. In the case of the daemon it was
3019 behaving confusingly. When first started, it used values for whoever
3020 started the daemon, but after a SIGHUP it used the Exim user (because it
3021 calls itself on a restart). I have changed the code so that it now always
3024 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
3025 message are rejected with the same error (e.g. no authentication or bad
3026 sender address), and a DATA command is nevertheless sent (as can happen
3027 with PIPELINING or a stupid MUA), the error message that was given to the
3028 RCPT commands is included in the rejection of the DATA command. This is
3029 intended to be helpful for MUAs that show only the final error to their
3032 PH/42 Another patch from the Sieve maintainer.
3034 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
3035 Eximstats - Fixed some broken HTML links and added missing column headers
3037 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
3038 columns for Rejects, Temp Rejects, Ham, and Spam rows.
3040 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
3042 PH/43 Yet another patch from the Sieve maintainer.
3044 PH/44 I found a way to check for a TCP/IP connection going away before sending
3045 the response to the final '.' that terminates a message, but only in the
3046 case where the client has not sent further data following the '.'
3047 (unfortunately, this is allowed). However, in many cases there won't be
3048 any further data because there won't be any more messages to send. A call
3049 to select() can be used: if it shows that the input is "ready", there is
3050 either input waiting, or the socket has been closed. An attempt to read
3051 the next input character can distinguish the two cases. Previously, Exim
3052 would have sent an OK response which the client would never have see.
3053 This could lead to message repetition. This fix should cure that, at
3054 least in a lot of common cases.
3056 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
3057 advertised in response to EHLO.
3063 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
3064 fixed by 4.65/MH/01 (is this a record?) are fixed:
3066 (i) An empty string was always treated as zero by the numeric comparison
3067 operators. This behaviour has been restored.
3069 (ii) It is documented that the numeric comparison operators always treat
3070 their arguments as decimal numbers. This was broken in that numbers
3071 starting with 0 were being interpreted as octal.
3073 While fixing these problems I realized that there was another issue that
3074 hadn't been noticed. Values of message_size_limit (both the global option
3075 and the transport option) were treated as octal if they started with 0.
3076 The documentation was vague. These values are now always treated as
3077 decimal, and I will make that clear in the documentation.
3083 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
3084 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
3087 MH/01 Don't check that the operands of numeric comparison operators are
3088 integers when their expansion is in "skipping" mode (fixes bug
3089 introduced by 4.64-PH/07).
3091 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
3092 child addresses, Exim now panics and dies. Previously, because the count
3093 is held in a short int, deliveries were likely to be lost. As such a
3094 large number of recipients for a single message is ridiculous
3095 (performance will be very, very poor), I have chosen to impose a limit
3096 rather than extend the field.
3102 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
3103 leftover -K file (the existence of which was triggered by #402).
3104 While we were at it, introduced process PID as part of the -K
3105 filename. This should rule out race conditions when creating
3108 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
3109 processing considerably. Previous code took too long for large mails,
3110 triggering a timeout which in turn triggers #401.
3112 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
3113 in the DK code in transports.c. sendfile() is not really portable,
3114 hence the _LINUX specificness.
3116 TF/01 In the add_headers option to the mail command in an Exim filter,
3117 there was a bug that Exim would claim a syntax error in any
3118 header after the first one which had an odd number of characters
3121 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
3122 callout verification, Exim cached a "reject" for the entire domain. This
3123 is correct for most verifications, but it is not correct for a recipient
3124 verification with use_sender or use_postmaster set, because in that case
3125 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
3126 case of MAIL FROM:<> rejection from other early rejections (e.g.
3127 rejection of HELO). When verifying a recipient using a non-null MAIL
3128 address, the cache is ignored if it shows MAIL FROM:<> rejection.
3129 Whatever the result of the callout, the value of the domain cache is
3130 left unchanged (for any other kind of callout, getting as far as trying
3131 RCPT means that the domain itself is ok).
3133 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
3136 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
3137 manifest itself as EPIPE rather than ECONNECT. When tidying away a
3138 session, the daemon ignores ECONNECT errors and logs others; it now
3139 ignores EPIPE as well.
3141 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
3142 (quoted-printable decoding).
3144 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
3145 later the small subsequent patch to fix an introduced bug.
3147 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
3149 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
3151 PH/08 An error is now given if message_size_limit is specified negative.
3153 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
3154 to be given (somewhat) arbitrary names.
3156 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
3159 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
3160 miscellaneous code fixes
3162 PH/10 Added the log_reject_target ACL modifier to specify where to log
3165 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
3166 hostname. This is wrong, because it relates to the incoming message (and
3167 probably the interface on which it is arriving) and not to the outgoing
3168 callout (which could be using a different interface). This has been
3169 changed to use the value of the helo_data option from the smtp transport
3170 instead - this is what is used when a message is actually being sent. If
3171 there is no remote transport (possible with a router that sets up host
3172 addresses), $smtp_active_hostname is used.
3174 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
3175 tweaks were necessary in order to get it to work (see also 21 below):
3176 (a) The code assumed that strncpy() returns a negative number on buffer
3177 overflow, which isn't the case. Replaced with Exim's string_format()
3179 (b) There were several signed/unsigned issues. I just did the minimum
3180 hacking in of casts. There is scope for a larger refactoring.
3181 (c) The code used strcasecmp() which is not a standard C function.
3182 Replaced with Exim's strcmpic() function.
3183 (d) The code set only $1; it now sets $auth1 as well.
3184 (e) A simple test gave the error "authentication client didn't specify
3185 service in request". It would seem that Dovecot has changed its
3186 interface. Fortunately there's a specification; I followed it and
3187 changed what the client sends and it appears to be working now.
3189 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
3192 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
3193 address A is aliased to B and C, where B exists and C does not. Without
3194 -v the output is "A verified" because verification stops after a
3195 successful redirection if more than one address is generated. However,
3196 with -v the child addresses are also verified. Exim was outputting "A
3197 failed to verify" and then showing the successful verification for C,
3198 with its parentage. It now outputs "B failed to verify", showing B's
3199 parentage before showing the successful verification of C.
3201 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
3202 look up a TXT record in a specific list after matching in a combined
3205 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
3206 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
3207 they consult the DNS. I had assumed they would set it the way they
3208 wanted; and indeed my experiments on Linux seem to show that in some
3209 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
3210 To be on the safe side, however, I have now made the interface to
3211 host_find_byname() similar to host_find_bydns(), with an argument
3212 containing the DNS resolver options. The host_find_byname() function now
3213 sets these options at its start, just as host_find_bydns() does. The smtp
3214 transport options dns_qualify_single and dns_search_parents are passed to
3215 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
3216 of host_find_byname() use the default settings of RES_DEFNAMES
3217 (qualify_single) but not RES_DNSRCH (search_parents).
3219 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
3220 spool_read_header() do less string testing, by means of a preliminary
3221 switch on the second character of optional "-foo" lines. (This is
3222 overdue, caused by the large number of possibilities that now exist.
3223 Originally there were few.) While I was there, I also converted the
3224 str(n)cmp tests so they don't re-test the leading "-" and the first
3225 character, in the hope this might squeeze out yet more improvement.
3227 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
3228 flag allowing group syntax was set by the header_syntax check but not
3229 turned off, possible causing trouble later; (2) The flag was not being
3230 set at all for the header_verify test, causing "group"-style headers to
3231 be rejected. I have now set it in this case, and also caused header_
3232 verify to ignore an empty address taken from a group. While doing this, I
3233 came across some other cases where the code for allowing group syntax
3234 while scanning a header line wasn't quite right (mostly, not resetting
3235 the flag correctly in the right place). These bugs could have caused
3236 trouble for malformed header lines. I hope it is now all correct.
3238 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
3239 with the "reply" argument non-NULL. The code, however (which originally
3240 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
3241 but it didn't always do it. This confused somebody who was copying the
3242 code for some other use. I have removed all the tests.
3244 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
3245 feature that was used to support insecure browsers during the U.S. crypto
3246 embargo. It requires special client support, and Exim is probably the
3247 only MTA that supported it -- and would never use it because real RSA is
3248 always available. This code has been removed, because it had the bad
3249 effect of slowing Exim down by computing (never used) parameters for the
3250 RSA_EXPORT functionality.
3252 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
3253 authenticator to fail if there's a tab character in the incoming data
3254 (there should never be unless someone is messing about, as it's supposed
3255 to be base64-encoded). Also added, on Timo's advice, the "secured" option
3256 if the connection is using TLS or if the remote IP is the same as the
3257 local IP, and the "valid-client-cert option" if a client certificate has
3260 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
3261 authenticators. This can be used for authorization after authentication
3262 succeeds. (In the case of plaintext, it servers for both authentication
3265 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
3266 if any retry times were supplied.
3268 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
3269 connection, where there is no client IP address to check. In this
3270 situation, the verify now always succeeds.
3272 PH/25 Applied John Jetmore's -Mset patch.
3274 PH/26 Added -bem to be like -Mset, but loading a message from a file.
3276 PH/27 In a string expansion for a processed (not raw) header when multiple
3277 headers of the same name were present, leading whitespace was being
3278 removed from all of them, but trailing whitespace was being removed only
3279 from the last one. Now trailing whitespace is removed from each header
3280 before concatenation. Completely empty headers in a concatenation (as
3281 before) are ignored.
3283 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
3284 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
3286 PH/29 [Removed. This was a change that I later backed out, and forgot to
3287 correct the ChangeLog entry (that I had efficiently created) before
3288 committing the later change.]
3290 PH/30 Exim was sometimes attempting to deliver messages that had suffered
3291 address errors (4xx response to RCPT) over the same connection as other
3292 messages routed to the same hosts. Such deliveries are always "forced",
3293 so retry times are not inspected. This resulted in far too many retries
3294 for the affected addresses. The effect occurred only when there were more
3295 hosts than the hosts_max_try setting in the smtp transport when it had
3296 the 4xx errors. Those hosts that it had tried were not added to the list
3297 of hosts for which the message was waiting, so if all were tried, there
3298 was no problem. Two fixes have been applied:
3300 (i) If there are any address or message errors in an SMTP delivery, none
3301 of the hosts (tried or untried) are now added to the list of hosts
3302 for which the message is waiting, so the message should not be a
3303 candidate for sending over the same connection that was used for a
3304 successful delivery of some other message. This seems entirely
3305 reasonable: after all the message is NOT "waiting for some host".
3306 This is so "obvious" that I'm not sure why it wasn't done
3307 previously. Hope I haven't missed anything, but it can't do any
3308 harm, as the worst effect is to miss an optimization.
3310 (ii) If, despite (i), such a delivery is accidentally attempted, the
3311 routing retry time is respected, so at least it doesn't keep
3312 hammering the server.
3314 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
3315 in ${readsocket because some servers need this prod.
3317 PH/32 Added some extra debug output when updating a wait-xxx database.
3319 PH/33 The hint "could be header name not terminated by colon", which has been
3320 given for certain expansion errors for a long time, was not being given
3321 for the ${if def:h_colon_omitted{... case.
3323 PH/34 The spec says: "With one important exception, whenever a domain list is
3324 being scanned, $domain contains the subject domain." There was at least
3325 one case where this was not true.
3327 PH/35 The error "getsockname() failed: connection reset by peer" was being
3328 written to the panic log as well as the main log, but it isn't really
3329 panic-worthy as it just means the connection died rather early on. I have
3330 removed the panic log writing for the ECONNRESET error when getsockname()
3333 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
3334 runs only) independently of the message's sender address. This meant
3335 that, if the 4xx error was in fact related to the sender, a different
3336 message to the same recipient with a different sender could confuse
3337 things. In particular, this can happen when sending to a greylisting
3338 server, but other circumstances could also provoke similar problems.
3339 I have changed the default so that the retry time for these errors is now
3340 based a combination of the sender and recipient addresses. This change
3341 can be overridden by setting address_retry_include_sender=false in the
3344 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
3345 remote server are returned as part of bounce messages. This was not
3346 happening for LMTP over a pipe (the lmtp transport), but now it is the
3347 same for both kinds of LMTP.
3349 PH/38 Despite being documented as not happening, Exim was rewriting addresses
3350 in header lines that were in fact CNAMEs. This is no longer the case.
3352 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
3353 and queue runs started by the daemon processed all messages. This has
3354 been fixed so that -R and -S can now usefully be given with -q<time>.
3356 PH/40 Import PCRE release 6.7 (fixes some bugs).
3358 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
3360 PH/42 Give an error if -q is specified more than once.
3362 PH/43 Renamed the variables $interface_address and $interface_port as
3363 $received_ip_address and $received_port, to make it clear that these
3364 values apply to message reception, and not to the outgoing interface when
3365 a message is delivered. (The old names remain recognized, of course.)
3367 PH/44 There was no timeout on the connect() call when using a Unix domain
3368 socket in the ${readsocket expansion. There now is.
3370 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
3371 be meaningful with "accept".
3373 SC/01 Eximstats V1.43
3374 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
3376 SC/02 Eximstats V1.44
3377 Use a glob alias rather than an array ref in the generated
3378 parser. This improves both readability and performance.
3380 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
3381 Collect SpamAssassin and rejection statistics.
3382 Don't display local sender or destination tables unless
3383 there is data to show.
3384 Added average volumes into the top table text output.
3386 SC/04 Eximstats V1.46
3387 Collect data on the number of addresses (recipients)
3388 as well as the number of messages.
3390 SC/05 Eximstats V1.47
3391 Added 'Message too big' to the list of mail rejection
3392 reasons (thanks to Marco Gaiarin).
3394 SC/06 Eximstats V1.48
3395 Mainlog lines which have GMT offsets and are too short to
3396 have a flag are now skipped.
3398 SC/07 Eximstats V1.49 (Alain Williams)
3399 Added the -emptyok flag.
3401 SC/08 Eximstats V1.50
3402 Fixes for obtaining the IP address from reject messages.
3404 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
3405 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
3406 whitespace changes from 4.64-PH/27
3408 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
3411 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
3412 are found, allow negative numbers in numeric criteria)
3414 JJ/06 exipick.20061117.2, added new $message_body_missing variable
3416 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
3417 to match changes made in 4.64-PH/43
3419 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
3421 PH/47 Put in an explicit test for a DNS lookup of an address record where the
3422 "domain" is actually an IP address, and force a failure. This locks out
3423 those revolvers/nameservers that support "A-for-A" lookups, in
3424 contravention of the specifications.
3426 PH/48 When a host name was looked up from an IP address, and the subsequent
3427 forward lookup of the name timed out, the host name was left in
3428 $sender_host_name, contrary to the specification.
3430 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
3431 restricted to single-key lookups, Exim was not diagnosing an error if
3432 * or *@ was used with a query-style lookup.
3434 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
3436 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
3437 long ago, but noone interested enough thought of it. Let's just say that
3438 the "1.1" means that there are some new functions that weren't there at
3439 some point in the past.
3441 PH/51 Error processing for expansion failure of helo_data from an smtp
3442 transport during callout processing was broken.
3444 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
3445 tested/used via the -bh/-bhc/-bs options.
3447 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
3448 bug, fixed in subsequent PCRE releases).
3450 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
3451 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
3453 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
3459 SC/01 Use a glob alias rather than an array ref in eximstats generated
3460 parser. This improves both readability and performance.
3462 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
3463 Don't display local sender or destination tables in eximstats unless
3464 there is data to show.
3465 Added average volumes into the eximstats top table text output.
3467 SC/03 Collect data on the number of addresses (recipients) as well
3468 as the number of messages in eximstats.
3470 TF/01 Correct an error in the documentation for the redirect router. Exim
3471 does (usually) call initgroups() when daemonizing.
3473 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
3474 with consistent privilege compared to when running as a daemon.
3476 TF/03 Note in the spec that $authenticated_id is not set for local
3477 submissions from trusted users.
3479 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
3480 Thanks to Dean Brooks <dean@iglou.com> for the patch.
3482 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
3483 by adding some example configuration directives to the default
3484 configuration file. A little bit of work is required to uncomment the
3485 directives and define how usernames and passwords are checked, but
3486 there is now a framework to start from.
3488 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
3489 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
3490 without this. I don't know how relevant this is to other LDAP libraries.
3492 PH/02 Add the verb name to the "unknown ACL verb" error.
3494 PH/03 Magnus Holmgren's patch for filter_prepend_home.
3496 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
3498 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
3499 directory not expanded when it should be if an expanded home directory
3500 was set for the address (which is overridden by the transport).
3502 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3505 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3506 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3507 because it is too late at that time, and has no effect.
3509 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3510 security issue with \' (bugzilla #107). I could not use the
3511 PQescapeStringConn() function, because it needs a PGconn value as one of
3514 PH/08 When testing addresses using -bt, indicate those final addresses that
3515 are duplicates that would not cause an additional delivery. At least one
3516 person was confused, thinking that -bt output corresponded to deliveries.
3517 (Suppressing duplicates isn't a good idea as you lose the information
3518 about possibly different redirections that led to the duplicates.)
3520 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3521 systems where poll() doesn't work, in particular OS X.
3523 PH/10 Added more information to debugging output for retry time not reached.
3525 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3526 operations in malware.c.
3528 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3531 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3532 syslog_duplication set false, log lines that would normally be written
3533 both the the main log and to the reject log were not written to syslog at
3536 PH/14 In the default configuration, change the use of "message" in ACL warn
3537 statements to "add_header".
3539 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3540 not followed by a command (e.g. "seen endif").
3542 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3543 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3546 PH/17 Added extra conditions to the default value of delay_warning_condition
3550 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3551 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3552 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3555 The Auto-Submitted: and various List- headers are standardised, whereas I
3556 don't think Precedence: ever was.
3558 PH/18 Refactored debugging code in route_finduser() to show more information,
3559 in particular, the error code if getpwnam() issues one.
3561 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3562 This is apparently needed in addition to the PH/07 change above to avoid
3563 any possible encoding problems.
3565 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3566 but not after initializing Perl.
3568 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3569 output them only if debugging. By default they are written stderr,
3570 apparently, which is not desirable.
3572 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3575 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3578 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3580 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3581 authenticated or an ident call has been made. Suppress the default
3582 values for $authenticated_id and $authenticated_sender (but permit -oMai
3583 and -oMas) when testing with -bh.
3585 PH/24 Re-jigged the order of the tests in the default configuration so that the
3586 tests for valid domains and recipients precede the DNS black list and CSA
3587 tests, on the grounds that those ones are more expensive.
3589 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3590 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3591 This bug exists in every version of Exim that I still have, right back to
3594 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3595 However, an attempt to turn on case-sensitivity in a regex key by
3596 including (?-i) didn't work because the subject string was already
3597 lowercased, and the effects were non-intuitive. It turns out that a
3598 one-line patch can be used to allow (?-i) to work as expected.
3604 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3605 other effects) broke the use of negated acl sub-conditions.
3607 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3610 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3611 "Deny" causes Exim to reject the incoming connection with a 554 error.
3612 Unfortunately, if there is a major crisis, such as a disk failure,
3613 tcp-wrappers gives "deny", whereas what one would like would be some
3614 kind of temporary error. A kludge has been added to help with this.
3615 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3616 554 error is used if errno is still zero or contains ENOENT (which occurs
3617 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3620 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3622 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3623 errors. Otherwise a message that provokes a temporary error (when other
3624 messages do not) can cause a whole host to time out.
3626 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3627 addresses were routed directly to files or pipes from a redirect router.
3628 File deliveries just didn't batch; pipe deliveries might have suffered
3631 PH/06 A failure to get a lock for a hints database would erroneously always say
3632 "Failed to get write lock", even when it was really a read lock.
3634 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3635 of 0600. This has been changed to use the value of the lockfile_mode
3636 option (which defaults to 0600).
3638 PH/08 Applied small patch from the Sieve maintainer.
3640 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3641 folder from quota calculations, a direct delivery into this folder messed
3642 up the contents of the maildirsize file. This was because the regex was
3643 used only to exclude .Trash (or whatever) when the size of the mailbox
3644 was calculated. There was no check that a delivery was happening into an
3645 excluded directory. This bug has been fixed by ignoring all quota
3646 processing for deliveries into excluded directories.
3648 PH/10 Added the maildirfolder_create_regex option to appendfile.
3654 PH/01 The code for finding all the local interface addresses on a FreeBSD
3655 system running IPv6 was broken. This may well have applied to all BSD
3656 systems, as well as to others that have similar system calls. The broken
3657 code found IPv4 interfaces correctly, but gave incorrect values for the
3658 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3659 that it would not match correctly against @[] and not recognize the IPv6
3662 PH/02 The ipliteral router was not recognizing addresses of the form user@
3663 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3665 PH/03 Added disable_ipv6.
3667 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3668 decoded form, because it is most often used to construct To: headers
3669 lines in autoreplies, and the decoded form may well be syntactically
3670 invalid. However, $reply_address has leading white space removed, and all
3671 newlines turned into spaces so that the autoreply transport does not
3674 PH/05 If group was specified without a user on a router, and no group or user
3675 was specified on a transport, the group from the router was ignored.
3677 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3678 for visible compile-time settings that can be used to change these
3679 numbers, for those that want even more. Backwards compatibility with old
3680 spool files has been maintained. However, going back to a previous Exim
3681 release will lost any variables that are in spool files.
3683 PH/07 Two small changes when running in the test harness: increase delay when
3684 passing a TCP/IP connection to a new process, in case the original
3685 process has to generate a bounce, and remove special handling of
3686 127.0.0.2 (sic), which is no longer necessary.
3688 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3689 be the same on different OS.
3691 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3694 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3695 whether --show-vars was specified or not
3697 JJ/02 exipick: Added support for new ACL variable spool format introduced
3700 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3701 syntactically invalid From: or Reply-to: line, and a filter used this to
3702 generate an autoreply, and therefore failed to obtain an address for the
3703 autoreply, Exim could try to deliver to a non-existent relative file
3704 name, causing unrelated and misleading errors. What now happens is that
3705 it logs this as a hard delivery error, but does not attempt to create a
3708 PH/11 The exinext utility has a -C option for testing purposes, but although
3709 the given file was scanned by exinext itself; it wasn't being passed on
3710 when Exim was called.
3712 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3713 an end-of-file indication when reading a command response.
3715 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3716 compiled. In many other places in Exim, IPv6 addresses are always
3717 recognized, so I have changed this. It also means that IPv4 domain
3718 literals of the form [IPV4:n.n.n.n] are now always recognized.
3720 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3721 used if the router is not running as root, for example, when verifying at
3722 ACL time, or when using -bh. The debugging output from this situation was
3723 non-existent - all you got was a failure to exec. I have made two
3726 (a) Failures to set uid/gid, the current directory, or a process leader
3727 in a subprocess such as that created by queryprogram now generate
3728 suitable debugging output when -d is set.
3730 (b) The queryprogram router detects when it is not running as root,
3731 outputs suitable debugging information if -d is set, and then runs
3732 the subprocess without attempting to change uid/gid.
3734 PH/15 Minor change to Makefile for building test_host (undocumented testing
3737 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3738 additional section of a DNS packet that returns MX or SRV records.
3739 Instead, it always explicitly searches for A/AAAA records. This avoids
3740 major problems that occur when a DNS server includes only records of one
3741 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3742 fixed another bug: if SRV records were looked up and the corresponding
3743 address records were *not* found in the additional section, the port
3744 values from the SRV records were lost.
3746 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3747 using the correct key (the original address) when searching the retry
3748 rules in order to find which one to use for generating the retry hint.
3750 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3751 adding the default one. Similarly, if it contains a Reply-To: header, the
3752 errors_reply_to option, if set, is not used.
3754 PH/19 When calculating a retry time, Exim used to measure the "time since
3755 failure" by looking at the "first failed" field in the retry record. Now
3756 it does not use this if it is later than than the arrival time of the
3757 message. Instead it uses the arrival time. This makes for better
3758 behaviour in cases where some deliveries succeed, thus re-setting the
3759 "first failed" field. An example is a quota failure for a huge message
3760 when small messages continue to be delivered. Without this change, the
3761 "time since failure" will always be short, possible causing more frequent
3762 delivery attempts for the huge message than are intended.
3763 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3765 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3766 $1, $2, $3) because the numerical variables can be reset during some
3767 expansion items (e.g. "match"), thereby losing the authentication data.
3769 PH/21 Make -bV show the size of off_t variables so that the test suite can
3770 decide whether to run tests for quotas > 2G.
3772 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3773 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3774 filecount value is greater than 2G or if a quota value is greater than 2G
3775 on a system where the size of off_t is not greater than 4, a panic error
3778 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3779 never match. The debug and -bh output now contains an explicit error
3780 message indicating a malformed IPv4 address or mask.
3782 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3783 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3784 PH/23 above applies.
3786 PH/25 Do not write to syslog when running in the test harness. The only
3787 occasion when this arises is a failure to open the main or panic logs
3788 (for which there is an explicit test).
3790 PH/26 Added the /no_tell option to "control=freeze".
3792 PH/27 If a host name lookup failed very early in a connection, for example, if
3793 the IP address matched host_lookup and the reverse lookup yielded a name
3794 that did not have a forward lookup, an error message of the form "no IP
3795 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3796 could be logged. Now it outputs the IP address instead of "NULL".
3798 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3799 allows the sender and the authenticated sender to be set when
3800 submitting a message from within Exim. Since child_open_exim() is
3801 documented for local_scan(), the new function should be too.
3803 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3804 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3805 results in an empty string is now treated as unset.
3807 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3809 PH/31 Added sender_verify_fail logging option.
3811 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3812 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3813 tidied the source and removed it altogether.
3815 PH/33 When a queue run was abandoned because the load average was too high, a
3816 log line was always written; now it is written only if the queue_run log
3817 selector is set. In addition, the log line for abandonment now contains
3818 information about the queue run such as the pid. This is always present
3819 in "start" and "stop" lines but was omitted from the "abandon" line.
3821 PH/34 Omit spaces between a header name and the colon in the error message that
3822 is given when verify = headers_syntax fails (if there are lots of them,
3823 the message gets confusing).
3825 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3826 names, as there are now some PTR records that contain slashes. This check
3827 is only to protect against broken name servers that fall over on strange
3828 characters, so the fact that it applies to all lookups doesn't matter.
3830 PH/36 Now that the new test suite is complete, we can remove some of the
3831 special code in Exim that was needed for the old test suite. For example,
3832 sorting DNS records because real resolvers return them in an arbitrary
3833 order. The new test suite's fake resolver always returns records in the
3836 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3837 bounces) except when queue_only is set, to avoid logging races between
3838 the different processes.
3840 PH/38 Panic-die if .include specifies a non-absolute path.
3842 PH/39 A tweak to the "H" retry rule from its user.
3844 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3845 a label. They prevented compilation on older perls.
3847 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3848 a warning to be raised on newish perls.
3850 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3851 on queue. Changes to match documented behaviour of showing count of
3852 messages matching specified criteria.
3854 PH/40 Changed the default ident timeout from 30s to 5s.
3856 PH/41 Added support for the use of login_cap features, on those BSD systems
3857 that have them, for controlling the resources used by pipe deliveries.
3859 PH/42 The content-scanning code uses fopen() to create files in which to put
3860 message data. Previously it was not paying any attention to the mode of
3861 the files. Exim runs with umask(0) because the rest of the code creates
3862 files with open(), and sets the required mode explicitly. Thus, these
3863 files were ending up world-writeable. This was not a big issue, because,
3864 being within the spool directory, they were not world-accessible. I have
3865 created a function called modefopen, which takes an additional mode
3866 argument. It sets umask(777), creates the file, chmods it to the required
3867 mode, then resets the umask. All the relevant calls to fopen() in the
3868 content scanning code have been changed to use this function.
3870 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3871 to 24 hours. This avoids potential overflow problems when processing G
3872 and H retry rules. I suspect nobody ever tinkers with this value.
3874 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3876 PH/45 When the plaintext authenticator is running as a client, the server's
3877 challenges are checked to ensure they are valid base64 strings. By
3878 default, the authentication attempt is cancelled if an invalid string is
3879 received. Setting client_ignore_invalid_base64 true ignores these errors.
3880 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3881 they are received. Thus, the responses can be made to depend on the
3882 challenges. If an invalid string is ignored, an empty string is placed in
3885 PH/46 Messages that are created by the autoreply transport now contains a
3886 References: header, in accordance with RFCs 2822 and 3834.
3888 PH/47 Added authenticated_sender_force to the smtp transport.
3890 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3892 PH/49 Installed latest patch from the Sieve maintainer.
3894 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3895 was also set, the appendfile transport was unnecessarily scanning a
3896 directory of message files (e.g. for maildir delivery) to find the count
3897 of files (along with the size), even though it did not need this
3898 information. It now does the scan only if it needs to find either the
3899 size of the count of files.
3901 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3903 PH/52 Two bugs concerned with error handling when the smtp transport is
3906 (i) Exim was not creating retry information for temporary errors given
3907 for individual recipients after the DATA command when the smtp transport
3908 was used in LMTP mode. This meant that they could be retried too
3909 frequently, and not timed out correctly.
3911 (ii) Exim was setting the flag that allows error details to be returned
3912 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3913 recipients that were returned after the DATA command.
3915 PH/53 This is related to PH/52, but is more general: for any failing address,
3916 when detailed error information was permitted to be returned to the
3917 sender, but the error was temporary, then after the final timeout, only
3918 "retry timeout exceeded" was returned. Now it returns the full error as
3919 well as "retry timeout exceeded".
3921 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3922 do this, and (what is worse) MTAs that accept it.
3924 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3925 will now be deprecated.
3927 PH/56 New os.c-cygwin from the Cygwin maintainer.
3929 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3930 formats (previously only available in exim formats via -bpr, -bpru,
3931 and -bpra. Now also available in native and exiqgrep formats)
3933 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3934 with very large, slow to parse queues
3936 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3938 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3940 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3941 responses to authentication challenges, though it was showing the
3942 challenges; (ii) I've removed the CR characters from the debug output for
3945 PH/58 Allow for the insertion of a newline as well as a space when a string
3946 is turned into more than one encoded-word during RFC 2047 encoding. The
3947 Sieve code now uses this.
3949 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3950 data_4xx, lost_connection, tls_required.
3952 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3953 message was being sent as an SMTP response.
3955 PH/61 Add -l and -k options to exicyclog.
3957 PH/62 When verifying, if an address was redirected to one new address, so that
3958 verification continued, and the new address failed or deferred after
3959 having set something in $address_data, the value of $address_data was not
3960 passed back to the ACL. This was different to the case when no
3961 redirection occurred. The value is now passed back in both cases.
3963 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3964 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3965 use login_cap.h, so on its own it isn't the distinguishing feature. The
3966 new name refers directly to the setclassresources() function.
3968 PH/65 Added configuration files for NetBSD3.
3970 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3972 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3973 is preferred over IPv4.
3975 PH/68 The bounce_return_message and bounce_return_body options were not being
3976 honoured for bounces generated during the reception of non-SMTP messages.
3977 In particular, this applied to messages rejected by the ACL. This bug has
3978 been fixed. However, if bounce_return_message is true and bounce_return_
3979 body is false, the headers that are returned for a non-SMTP message
3980 include only those that have been read before the error was detected.
3981 (In the case of an ACL rejection, they have all been read.)
3983 PH/69 The HTML version of the specification is now built in a directory called
3984 spec_html instead of spec.html, because the latter looks like a path with
3985 a MIME-type, and this confuses some software.
3987 PH/70 Catch two compiler warnings in sieve.c.
3989 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3990 function verify_get_ident() calls ip_connect() to connect a socket, but
3991 if the "connect()" function timed out, ip_connect() used to close the
3992 socket. However, verify_get_ident() also closes the socket later, and in
3993 between Exim writes to the log, which may get opened at this point. When
3994 the socket was closed in ip_connect(), the log could get the same file
3995 descriptor number as the socket. This naturally causes chaos. The fix is
3996 not to close the socket in ip_connect(); the socket should be closed by
3997 the function that creates it. There was only one place in the code where
3998 this was missing, in the iplookup router, which I don't think anybody now
3999 uses, but I've fixed it anyway.
4001 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
4002 well as to direct DNS lookups. Otherwise the handling of names in host
4003 lists is inconsistent and therefore confusing.
4009 PH/01 Two changes to the default runtime configuration:
4011 (1) Move the checks for relay_from_hosts and authenticated clients from
4012 after to before the (commented out) DNS black list checks.
4014 (2) Add control=submission to the relay_from_hosts and authenticated
4015 clients checks, on the grounds that messages accepted by these
4016 statements are most likely to be submissions.
4018 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
4020 (1) Generate an error if the third argument for the ${prvs expansion is
4023 (2) Treat a missing third argument of ${prvscheck as if it were an empty
4026 (3) Reset the variables that are obtained from the first argument of
4027 ${prvscheck and used in the second argument before leaving the code,
4028 because their memory is reclaimed, so using them afterwards may do
4031 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
4032 one (it's much easier than Tom thought :-).
4034 (5) Because of (4), we can now allow for the use of $prvscheck_result
4035 inside the third argument.
4037 PH/03 For some reason, the default setting of PATH when running a command from
4038 a pipe transport was just "/usr/bin". I have changed it to
4041 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
4042 anything to be listed in the output from -bV.
4044 PH/05 When a filter generated an autoreply, the entire To: header line was
4045 quoted in the delivery log line, like this:
4047 => >A.N.Other <ano@some.domain> <original@ddress> ...
4049 This has been changed so that it extracts the operative address. There
4050 may be more than one such address. If so, they are comma-separated, like
4053 => >ano@some.domain,ona@other.domain <original@ddress> ...
4055 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
4056 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
4057 not being looked up in the rDNS to get a host name, Exim was showing the
4058 IP address twice in Received: lines, even though the IP addresses were
4059 identical. For example:
4061 Received: from [1.2.3.4] (helo=[1.2.3.4])
4063 However, if the real host name was known, it was omitting the HELO data
4064 if it matched the actual IP address. This has been tidied up so that it
4065 doesn't show the same IP address twice.
4067 PH/07 When both +timestamp and +memory debugging was on, the value given by
4068 $tod_xxx expansions could be wrong, because the tod_stamp() function was
4069 called by the debug printing, thereby overwriting the timestamp buffer.
4070 Debugging no longer uses the tod_stamp() function when +timestamp is set.
4072 PH/08 When the original message was included in an autoreply transport, it
4073 always said "this is a copy of the message, including all the headers",
4074 even if body_only or headers_only was set. It now gives an appropriate
4077 PH/09 Applied a patch from the Sieve maintainer which:
4079 o fixes some comments
4080 o adds the (disabled) notify extension core
4081 o adds some debug output for the result of if/elsif tests
4082 o points to the current vacation draft in the documentation
4083 and documents the missing references header update
4087 o fixes a bug in processing the envelope test (when testing
4088 multiple envelope elements, the last element determined the
4091 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
4092 Electronic Mail") by including:
4094 Auto-submitted: auto-generated
4096 in the messages that it generates (bounce messages and others, such as
4097 warnings). In the case of bounce messages for non-SMTP messages, there was
4098 also a typo: it was using "Auto_submitted" (underscore instead of
4099 hyphen). Since every message generated by Exim is necessarily in response
4100 to another message, thes have all been changed to:
4102 Auto-Submitted: auto-replied
4104 in accordance with these statements in the RFC:
4106 The auto-replied keyword:
4108 - SHOULD be used on messages sent in direct response to another
4109 message by an automatic process,
4111 - MUST NOT be used on manually-generated messages,
4113 - MAY be used on Delivery Status Notifications (DSNs) and Message
4114 Disposition Notifications (MDNs),
4116 - MUST NOT be used on messages generated by automatic or periodic
4117 processes, except for messages which are automatic responses to
4120 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
4121 to the default Received: header definition.
4123 PH/12 Added log selector acl_warn_skipped (default on).
4125 PH/13 After a successful wildlsearch lookup, discard the values of numeric
4126 variables because (a) they are in the wrong storage pool and (b) even if
4127 they were copied, it wouldn't work properly because of the caching.
4129 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
4130 checking when decoding. Apparently there are clients that generate
4131 overlong encoded strings. Why am I not surprised?
4133 PH/15 If the first argument of "${if match_address" was not empty, but did not
4134 contain an "@" character, Exim crashed. Now it writes a panic log message
4135 and treats the condition as false.
4137 PH/16 In autoreply, treat an empty string for "once" the same as unset.
4139 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
4140 extension "envelope-auth". The code is finished and in agreement with
4141 other implementations, but there is no documentation so far and in fact,
4142 nobody wrote the draft yet. This extension is currently #undef'ed, thus
4143 not changing the active code.
4145 Print executed "if" and "elsif" statements when debugging is used. This
4146 helps a great deal to understand what a filter does.
4148 Document more things not specified clearly in RFC3028. I had all this
4149 sorted out, when out of a sudden new issues came to my mind. Oops."
4151 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
4154 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
4155 canonical form (as documented). However, after a host name lookup from
4156 the IP address, check_host() was doing a simple string comparison with
4157 addresses acquired from the DNS when checking that the found name did
4158 have the original IP as one of its addresses. Since any found IPv6
4159 addresses are likely to be in abbreviated form, the comparison could
4160 fail. Luckily, there already exists a function for doing the comparison
4161 by converting both addresses to binary, so now that is used instead of
4162 the text comparison.
4164 PH/20 There was another similar case to PH/19, when a complete host name was
4165 given in a host list; looking up its IP address could give an abbreviated
4166 form, whereas the current host's name might or might not be abbreviated.
4167 The same fix has been applied.
4173 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
4174 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
4177 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
4178 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
4180 PH/03 Typo: missing ".o" in src/pcre/Makefile.
4182 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
4183 header line, restrict the check to what is listed in RFCs 2369 and 2929.
4184 Also, for "Auto-Submitted", treat anything other than "no" as
4185 non-personal, in accordance with RFC 3834. (Previously it treated
4186 anything starting "auto-" as non-personal.)
4188 TF/01 The control=submission/name=... option had a problem with syntax
4189 errors if the name included a slash character. The /name= option
4190 now slurps the rest of the string, so it can include any characters
4191 but it must come last in the list of options (after /sender_retain
4194 PH/05 Some modifications to the interface to the fake nameserver for the new
4202 TK/01 Added the "success_on_redirect" address verification option. See
4203 NewStuff for rationale and an example.
4205 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
4207 PH/02 Patch to exigrep to allow it to work on syslog lines.
4209 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
4210 fread() to skip over the body file's header line, because in Cygwin the
4211 header line is locked and is inaccessible.
4213 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
4214 co-exist for some time) to make it clear that it is the Exim ID that is
4215 referenced, not the Message-ID: header line.
4217 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
4218 string_format() function, because snprintf() does not exist on all
4221 PH/06 The use of forbid_filter_existstest now also locks out the use of the
4222 ${stat: expansion item.
4224 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
4225 protocol synchronization error", to keep the pedants happy.
4227 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
4228 well as for IRIX systems, when gcc is being used. See the host.c source
4231 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
4233 PH/10 Named domain lists were not working if used in a queue_smtp_domains
4236 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
4237 transport and to the smtp transport in LMTP mode.
4239 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
4241 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
4242 run a filter in a subprocess. This could lead to confusion in subsequent
4243 lookups in the parent process. There should also be a search_tidyup() at
4244 the end of the subprocess.
4246 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
4247 only if the host matched helo_try_verify_hosts, which caused the
4248 verification to occur when the EHLO/HELO command was issued. The ACL just
4249 tested the remembered result. Now, if a previous verification attempt has
4250 not happened, "verify = helo" does it there and then.
4252 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
4254 TK/03 Fix log output including CR from clamd.
4256 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
4257 exist provoked a memory error which could cause a segfault.
4259 PH/15 Installed PCRE 6.2
4261 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
4263 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
4264 of the problem. Specifically, suggested +O2 rather than +O1 for the
4267 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
4269 PH/20 If a delivery was routed to a non-standard port by means of an SRV
4270 record, the port was not correctly logged when the outgoing_port log
4271 selector was set (it logged the transort's default port).
4273 PH/21 Added support for host-specific ports to manualroute, queryprogram,
4274 fallback_hosts, and "hosts" in the smtp transport.
4276 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
4277 host errors such as "Connection refused".
4279 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
4280 authentication with radiusclient 0.4.9:
4282 - Error returned from rc_read_config was caught wrongly
4283 - Username/password not passed on to radius server due to wrong length.
4285 The presumption is that some radiusclient API changes for 4.51/PH/17
4286 were not taken care of correctly. The code is still untested by me (my
4287 Linux distribution still has 0.3.2 of radiusclient), but it was
4288 contributed by a Radius user.
4290 PH/24 When doing a callout, the value of $domain wasn't set correctly when
4291 expanding the "port" option of the smtp transport.
4293 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
4294 while reading a MIME header. Thanks to Tom Hughes for a patch.
4296 PH/24 Include config.h inside local_scan.h so that configuration settings are
4299 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
4300 that in an ACL for RCPT (for example), you can examine exactly what was
4303 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
4304 commands, but it was not correctly comparing the address with the actual
4305 client host address. Thus, it would show the EHLO address in Received:
4306 header lines when this was not necessary.
4308 PH/27 Added the % operator to ${eval:}.
4310 PH/28 Exim tries to create and chdir to its spool directory when it starts;
4311 it should be ignoring failures (because with -C, for example, it has lost
4312 privilege). It wasn't ignoring creation failures other than "already
4315 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
4318 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
4319 because an input line was too long, either on its own, or by virtue of
4320 too many continuations, the temporary file was not being removed, and the
4321 return code was incorrect.
4323 PH/31 Missing "BOOL" in function definition in filtertest.c.
4325 PH/32 Applied Sieve patches from the maintainer.
4327 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
4329 PH/33 Added "verify = not_blind".
4331 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
4332 Local/Makefile (with some defaults set). These are used in built scripts
4333 such as exicyclog, but they have never been used in the exim_install
4334 script (though there are many overriding facilities there). I have
4335 arranged that the exim_install script now takes note of these two
4338 PH/35 Installed configuration files for Dragonfly.
4340 PH/36 When a locally submitted message by a trusted user did not contain a
4341 From: header, and the sender address was obtained from -f or from an SMTP
4342 MAIL command, and the trusted user did not use -F to supply a sender
4343 name, $originator_name was incorrectly used when constructing a From:
4344 header. Furthermore, $originator_name was used for submission mode
4345 messages from external hosts without From: headers in a similar way,
4346 which is clearly wrong.
4348 PH/37 Added control=suppress_local_fixups.
4350 PH/38 When log_selector = +received_sender was set, and the addition of the
4351 sender made the log line's construction buffer exactly full, or one byte
4352 less than full, an overflow happened when the terminating "\n" was
4355 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
4356 when the result of a list match is failure because a DNS lookup failed.
4358 PH/40 RM_COMMAND is now used in the building process.
4360 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
4361 the "build-* directories that it finds.
4363 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
4364 address in a domain literal was a prefix of an interface address.
4366 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
4367 when verifying a sender address, unless rewrite_headers is false.
4369 PH/44 Wrote a long comment about why errors_to addresses are verified as
4370 recipients, not senders.
4372 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
4373 the ratelimit ACL was added.
4375 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
4377 PH/46 Added extra information about PostgreSQL errors to the error string.
4379 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
4380 avoiding the need to install special zones in a real server. This is
4381 backwards compatible; if it can't find the fake resolver, it drops back.
4382 Thus, both old and new test suites can be run.
4384 TF/02 Added util/ratelimit.pl
4386 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
4387 clock is set back in time.
4389 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
4390 Candler <B.Candler@pobox.com>.
4392 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
4393 OK for addresses that are the result of redirections.
4395 PH/48 A number of further additions for the benefit of the new test suite,
4396 including a fake gethostbyname() that interfaces to the fake DNS resolver
4399 TF/06 The fix for widen_domains has also been applied to qualify_single and
4400 search_parents which are the other dnslookup options that can cause
4403 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
4406 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
4408 TF/07 Exim produced the error message "an SRV record indicated no SMTP
4409 service" if it encountered an MX record with an empty target hostname.
4410 The message is now "an MX or SRV record indicated no SMTP service".
4412 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
4413 if the DNS of the sending site is misconfigured. This is quite a
4414 common situation. This change restores the behaviour of treating a
4415 helo verification defer as a failure.
4417 PH/51 If self=fail was set on a router, the bounce message did not include the
4418 actual error message.
4424 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
4426 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
4427 command itself ended in error, the underlying message about the transport
4428 filter timeout was being overwritten with the pipe command error. Now the
4429 underlying error message should be appended to the second error message.
4431 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
4433 PH/02 Reduce the amount of output that "make" produces by default. Full output
4434 can still be requested.
4436 PH/03 The warning log line about a condition test deferring for a "warn" verb
4437 was being output only once per connection, rather than after each
4438 occurrence (because it was using the same function as for successful
4439 "warn" verbs). This seems wrong, so I have changed it.
4441 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
4442 it should not have, which might have caused a crash in the right
4443 circumstances, but probably never did.
4445 PH/04 Installed a modified version of Tony Finch's patch to make submission
4446 mode fix the return path as well as the Sender: header line, and to
4447 add a /name= option so that you can make the user's friendly name appear
4450 TF/03 Added the control = fakedefer ACL modifier.
4452 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
4453 Mark Lowes for thorough testing.
4455 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
4457 TK/03 Merged latest SRS patch from Miles Wilton.
4459 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
4460 with the definition in sysexits.h (which is #included earlier).
4461 Fortunately, Exim does not actually use EX_OK. The code used to try to
4462 preserve the sysexits.h value, by assuming that macro definitions were
4463 scanned for macro replacements. I have been disabused of this notion,
4464 so now the code just undefines EX_OK before #including unistd.h.
4466 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
4467 in the smtp transport. When a block could not be written in a single
4468 write() function, the timeout was being re-applied to each part-write.
4469 This seems wrong - if the receiver was accepting one byte at a time it
4470 would take for ever. The timeout is now adjusted when this happens. It
4471 doesn't have to be particularly precise.
4473 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
4474 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
4476 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
4477 without a domain if the check to <postmaster@domain> fails.
4479 SC/01 Eximstats: added -xls and the ability to specify output files
4480 (patch written by Frank Heydlauf).
4482 SC/02 Eximstats: use FileHandles for outputting results.
4484 SC/03 Eximstats: allow any combination of xls, txt, and html output.
4486 SC/04 Eximstats: fixed display of large numbers with -nvr option
4488 SC/05 Eximstats: fixed merging of reports with empty tables.
4490 SC/06 Eximstats: added the -include_original_destination flag
4492 SC/07 Eximstats: removed tabs and trailing whitespace.
4494 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
4496 TK/06 MBOX spool code: Add real "From " MBOX separator line
4497 so the .eml file is really in mbox format (even though
4498 most programs do not really care). Patch from Alex Miller.
4500 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4501 The latter is generated from $received_to and is only set if the
4502 message has one envelope recipient. SA can use these headers,
4503 obviously out-of-the-box. Patch from Alex Miller.
4505 PH/08 The ${def test on a variable was returning false if the variable's
4506 value was "0", contrary to what the specification has always said!
4507 The result should be true unless the variable is empty.
4509 PH/09 The syntax error of a character other than { following "${if
4510 def:variable_name" (after optional whitespace) was not being diagnosed.
4511 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4512 accidental colon was present, for example, could give incorrect results.
4514 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4515 result is used (not including appendfile, where other changes are about
4518 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4519 This involved changing a lot of size variables from int to off_t. It
4520 should work with maildirs and everything.
4522 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4523 spamd dying while we are connected to it.
4525 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4528 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4529 function for BDB 4.3.
4531 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4533 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4534 the sources for PCRE have been re-arranged and more files are now
4537 PH/15 The code I had for printing potentially long long variables in PH/11
4538 above was not the best (it lost precision). The length of off_t variables
4539 is now inspected at build time, and an appropriate printing format (%ld
4540 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4541 to be "long long int" or "long int". This is needed for the internal
4542 formatting function string_vformat().
4544 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4545 the configuration file to be ":syslog", then the script "guesses" where
4546 the logs files are, rather than using the compiled in default. In our
4547 case the guess is not the same as the compiled default, so the script
4548 suddenly stopped working when I started to use syslog. The patch checks
4549 to see if log_file_path is "". If so, it attempts to read it from exim
4550 with no configuration file to get the compiled in version, before it
4551 falls back to the previous guessing code."
4553 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4554 implementing BATV in an Exim configuration. See NewStuff for the gory
4557 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4558 Makefile that are specific to HP-UX.
4560 PH/18 If the "use_postmaster" option was set for a recipient callout together
4561 with the "random" option, the postmaster address was used as the MAIL
4562 FROM address for the random test, but not for the subsequent recipient
4563 test. It is now used for both.
4565 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4566 patch removes a few documentation additions to RFC 3028, because the
4567 latest draft now contains them. It adds the new en;ascii-case comparator
4568 and a new error check for 8bit text in MIME parts. Comparator and
4569 require names are now matched exactly. I enabled the subaddress
4570 extension, but it is not well tested yet (read: it works for me)."
4572 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4573 rework some of the code of TK/09 above to avoid the hardwired use of
4574 "%lld" and "long long". Replaced the call to snprintf() with a call to
4577 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4578 records point to non-existent hosts", "retry timeout exceeded", and
4579 "retry time not reached for any host after a long failure period".
4581 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4582 experimental DomainKeys support:
4584 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4585 (2) On an error such as an illegally used "control", the wrong name for
4586 the control was given.
4588 These problems did NOT occur unless DomainKeys support was compiled.
4590 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4592 PH/24 Added ${if match_ip condition.
4594 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4595 databases so that it will be absolutely obvious if a crash occurs in the
4596 DB library. This is a regular occurrence (often caused by mis-matched
4599 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4600 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4601 picked up on a user's system that detects such things. There doesn't seem
4602 to be a gcc warning option for this - only an attribute that has to be
4603 put on the function's prototype. It seems that in Fedora Core 4 they have
4604 set this on a number of new functions. No doubt there will be more in due
4607 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4608 specify a transport. However, if an address that was verified by such a
4609 router was the subject of a callout, Exim crashed because it tried to
4610 read the rcpt_include_affixes from the non-existent transport. Now it
4611 just assumes that the setting of that option is false. This bug was
4612 introduced by 4.51/PH/31.
4614 PH/28 Changed -d+all to exclude +memory, because that information is very
4615 rarely of interest, but it makes the output a lot bigger. People tend to
4616 do -d+all out of habit.
4618 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4619 code in os-type was giving problems when libc.so lives in lib64, like on
4622 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4623 aren't the modern standard, and it seems that some systems' include files
4624 don't always have them. Exim was already checking for some of the newer
4625 ones like T_AAAA, and defining it itself. I've added checks for all the
4626 record types that Exim uses.
4628 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4629 not automatically generating a new one, as it is supposed to. This
4630 prevented TLS from working. If the file did exist, but contained invalid
4631 data, a new version was generated, as expected. It was only the case of a
4632 non-existent file that was broken.
4634 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4635 with a change in libdomainkeys > 0.64.
4637 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4638 from DNS. If the selector record carries the flag, it now has
4639 precedence over the domain-wide flag.
4641 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4643 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4644 the use of a port name that isn't defined in /etc/services) occurred, the
4645 message was deferred as in a normal delivery, and thus remained on the
4646 spool, instead of being failed because of the mua_wrapper setting. This
4647 is now fixed, and I tidied up some of the mua_wrapper messages at the
4650 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4651 the messages in a hash of arrays rather than using individual hashes.
4652 This is a bit cleaner and results in dramatic memory savings, albeit
4653 at a slight CPU cost.
4655 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4656 as requested by Marc Sherman.
4658 SC/10 Eximstats: added histograms for user specified patterns as requested
4661 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4663 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4664 fopen() in the content-scanning modules that did not already have it.
4670 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4671 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4673 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4675 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4677 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4678 to test Sieve filters that use "vacation".
4680 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4681 that changes the way the GnuTLS parameters are stored in the cache file.
4682 The new format can be generated externally. For backward compatibility,
4683 if the data in the cache doesn't make sense, Exim assumes it has read an
4684 old-format file, and it generates new data and writes a new file. This
4685 means that you can't go back to an older release without removing the
4688 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4689 work if there are any delivery delays because "one_time" forces the
4690 parent to be marked "delivered", so its unseen clone is never tried
4691 again. For this reason, Exim now forbids the simultaneous setting of
4694 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4695 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4696 entry if you want to know the details. The fix, however, neglected to
4697 consider the case when local delivery batching is involved. The test for
4698 "previously delivered" was not happening when checking to see if an
4699 address could be batched with a previous (undelivered) one; under
4700 certain circumstances this could lead to multiple deliveries to the same
4703 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4704 in its include files, and this causes problems building Exim.
4706 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4707 header_syntax) but Exim was just ignoring anything given after a slash.
4708 In particular, this caused confusion with an attempt to use "verify =
4709 reverse_host_lookup/defer_ok". An error is now given when options are
4710 supplied for verify items that do not have them. (Maybe reverse_host_
4711 lookup should have a defer_ok option, but that's a different point.)
4713 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4714 defined by RFC 821) to 2048, because there were problems with some AUTH
4715 commands, and RFC 1869 says the size should be increased for extended
4716 SMTP commands that take arguments.
4718 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4721 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4722 "unknown" error; now it says that the functionality isn't in the binary.
4724 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4725 an address' error message when a string expansion fails (syntax or
4726 whatever). Otherwise the password may appear in the log. Following change
4727 PH/42 below, there is no longer a chance of it appearing in a bounce
4730 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4732 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4733 of its addresses was ever tried. (Bugzilla bug #2.)
4735 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4736 the result incorrectly in the debug output. (It correctly added a newline
4737 to what was transported.)
4739 TF/01 Added $received_time.
4741 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4742 commented out examples of how to interface to a virus scanner and to
4743 SpamAssassin. Also added commented examples of av_scanner and
4744 spamd_address settings.
4746 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4747 and controls are allowed in which ACLs. There were a couple of minor
4748 errors. Some of the entries in the conditions table (which is a table of
4749 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4750 negation of where the condition IS allowed.
4752 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4754 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4755 header file does not have a version number, so I've had to invent a new
4756 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4757 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4758 radiusclient), but it was contributed by a Radius user.
4760 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4761 files or directories, for OpenSSL.
4763 PH/19 When an Exim process that is running as root has to create an Exim log
4764 file, it does so in a subprocess that runs as exim:exim so as to get the
4765 ownership right at creation (otherwise, other Exim processes might see
4766 the file with the wrong ownership). There was no test for failure of this
4767 fork() call, which would lead to the process getting stuck as it waited
4768 for a non-existent subprocess. Forks do occasionally fail when resources
4769 run out. I reviewed all the other calls to fork(); they all seem to check
4772 PH/20 When checking for unexpected SMTP input at connect time (before writing
4773 the banner), Exim was not dealing correctly with a non-positive return
4774 from the read() function. If the client had disconnected by this time,
4775 the result was a log entry for a synchronization error with an empty
4776 string after "input=" when read() returned zero. If read() returned -1
4777 (an event I could not check), uninitialized data bytes were printed.
4778 There were reports of junk text (parts of files, etc) appearing after
4781 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4783 PH/22 Added support for macro redefinition, and (re)definition in between
4784 driver and ACL definitions.
4786 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4787 forgetting to use the resulting value; it was using the unexpanded value.
4789 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4790 hadn't been configured. The fix is from Juergen Kreileder, who
4791 understands it better than I do:
4793 "Here's what I see happening with three configured cyrus_sasl
4794 authenticators configured (plain, login, cram-md5):
4796 On startup auth_cyrus_sasl_init() gets called for each of these.
4797 This means three calls to sasl_listmech() without a specified mech_list.
4798 => SASL tests which mechs of all available mechs actually work
4799 => three warnings about OTP not working
4800 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4802 With the patch, sasl_listmech() also gets called three times. But now
4803 SASL's mech_list option is set to the server_mech specified in the the
4804 authenticator. Or in other words, the answer from sasl_listmech()
4805 gets limited to just the mech you're testing for (which is different
4807 => the return list contains just 'plain' or 'login', 'cram-md5' or
4808 nothing depending on the value of ob->server_mech.
4810 I've just tested the patch: Authentication still works fine,
4811 unavailable mechs specified in the exim configuration are still
4812 caught, and the auth.log warnings about OTP are gone."
4814 PH/25 When debugging is enabled, the contents of the command line are added
4815 to the debugging output, even when log_selector=+arguments is not
4818 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4819 answer is "GNU", and only if the return is "GNU/something" is the answer
4822 PH/27 $acl_verify_message is now set immediately after the failure of a
4823 verification in an ACL, and so is available in subsequent modifiers. In
4824 particular, the message can be preserved by coding like this:
4826 warn !verify = sender
4827 set acl_m0 = $acl_verify_message
4829 Previously, $acl_verify_message was set only while expanding "message"
4830 and "log_message" when a very denied access.
4832 PH/28 Modified OS/os.c-Linux with
4834 -#ifndef OS_LOAD_AVERAGE
4835 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4837 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4838 nomenclature these days.)
4840 PH/29 Installed patch from the Sieve maintainer that adds the options
4841 sieve_useraddress and sieve_subaddress to the redirect router.
4843 PH/30 In these circumstances:
4844 . Two addresses routed to the same list of hosts;
4845 . First host does not offer TLS;
4846 . First host accepts first address;
4847 . First host gives temporary error to second address;
4848 . Second host offers TLS and a TLS session is established;
4849 . Second host accepts second address.
4850 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4851 and peerdn, if requested) that were in fact used only for the second
4854 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4855 attention to any local part prefix or suffix that was matched by the
4856 router that accepted the address. It now behaves in the same way as it
4857 does for delivery: the affixes are removed from the local part unless
4858 rcpt_include_affixes is set on the transport.
4860 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4861 timeout during the DATA phase of an incoming message.
4863 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4864 applied a patch sanctioned by the Sieve maintainer.
4866 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4867 the uid or gid is negative. A case of a negative gid caused this to be
4868 noticed. The fix allows for either to be negative.
4870 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4871 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4874 PH/36 The change PH/12 above was broken. Fixed it.
4876 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4877 the grounds that routing the same address twice would always produce the
4878 same answer. This might have been true once, but it is certainly no
4879 longer true now. Routing a child address may depend on the previous
4880 routing that produced that child. Some complicated redirection strategies
4881 went wrong when messages had multiple recipients, and made Exim's
4882 behaviour dependent on the order in which the addresses were given.
4884 I have moved the duplicate checking until after the routing is complete.
4885 Exim scans the addresses that are assigned to local and remote
4886 transports, and removes any duplicates. This means that more work will be
4887 done, as duplicates will always all be routed, but duplicates are
4888 presumably rare, so I don't expect this is of any significance.
4890 For deliveries to pipes, files, and autoreplies, the duplicate checking
4891 still happens during the routing process, since they are not going to be
4894 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4895 It corrects a timeout issue with spamd. This is Ian's comment: "The
4896 background is that sometimes spamd either never reads data from a
4897 connection it has accepted, or it never writes response data. The exiscan
4898 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4899 blindly assumes that writes won't block so it may never time out."
4901 PH/39 Allow G after quota size as well as K and M.
4903 PH/40 The value set for $authenticated_id in an authenticator may not contain
4904 binary zeroes or newlines because the value is written to log lines and
4905 to spool files. There was no check on this. Now the value is run through
4906 the string_printing() function so that such characters are converted to
4907 printable escape sequences.
4909 PH/41 $message_linecount is a new variable that contains the total number of
4910 lines in the message. Compare $body_linecount, which is the count for the
4913 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4914 bounce and delay warning messages, except in certain special cases, which
4917 (a) An SMTP error message from a remote host;
4918 (b) A message specified in a :fail: redirection;
4919 (c) A message specified in a "fail" command in a system filter;
4920 (d) A message specified in a FAIL return from the queryprogram router;
4921 (e) A message specified by the cannot_route_message router option.
4923 In these cases only, Exim does include the error details in bounce and
4924 warning messages. There are also a few cases where bland messages such
4925 as "unrouteable address" or "local delivery error" are given.
4927 PH/43 $value is now also set for the "else" part of a ${run expansion.
4929 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4930 being worked on, but at least Exim now implements the latest version to
4933 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4934 process to complete was treated as a delivery failure, a timeout while
4935 writing the message to the pipe was logged, but erroneously treated as a
4936 successful delivery. Such timeouts include transport filter timeouts. For
4937 consistency with the overall process timeout, these timeouts are now
4938 treated as errors, giving rise to delivery failures by default. However,
4939 there is now a new Boolean option for the pipe transport called
4940 timeout_defer, which, if set TRUE, converts the failures into defers for
4941 both kinds of timeout. A transport filter timeout is now identified in
4944 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4945 systems where "make" and "gmake" are different, calling "gmake" at top
4946 level broke things. I've arranged for the value of $(MAKE) to be passed
4947 from the Makefile to this script so that it can call the same version of
4951 A note about Exim versions 4.44 and 4.50
4952 ----------------------------------------
4954 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4955 changes of various kinds. As a consequence, a big documentation update was
4956 needed. This delayed the release for rather longer than seemed good, especially
4957 in the light of a couple of (minor) security issues. Therefore, the changes
4958 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4959 release. So 4.44 and 4.50 are in effect two different branches that both start
4962 I have left the 4.50 change log unchanged; it contains all the changes since
4963 4.43. The change log for 4.44 is below; many of its items are identical to
4964 those for 4.50. This seems to be the most sensible way to preserve the
4965 historical information.
4971 1. Minor wording change to the doc/README.SIEVE file.
4973 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4974 computation of the current number of files was incorrect.
4976 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4977 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4980 4. Give more explanation in the error message when the command for a transport
4981 filter fails to execute.
4983 5. There are several places where Exim runs a non-Exim command in a
4984 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4985 was being done only for the command run by the queryprogram router. It is
4986 now done for all such subprocesses. The other cases are: ${run, transport
4987 filters, and the commands run by the lmtp and pipe transports.
4989 6. Added CONFIGURE_GROUP build-time option.
4991 7. Some older OS have a limit of 256 on the maximum number of file
4992 descriptors. Exim was using setrlimit() to set 1000 as a large value
4993 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4994 systems. I've change it so that if it can't get 1000, it tries for 256.
4996 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4997 was an oversight, and furthermore, ever since the addition of extra
4998 controls (e.g. 4.43/32), the checks on when to allow different forms of
4999 "control" were broken. There should now be diagnostics for all cases when a
5000 control that does not make sense is encountered.
5002 9. Added the /retain_sender option to "control=submission".
5004 10. $recipients is now available in the predata ACL (oversight).
5006 11. Tidy the search cache before the fork to do a delivery from a message
5007 received from the command line. Otherwise the child will trigger a lookup
5008 failure and thereby defer the delivery if it tries to use (for example) a
5009 cached ldap connection that the parent has called unbind on.
5011 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
5012 of $address_data from the recipient verification was clobbered by the
5013 sender verification.
5015 13. The value of address_data from a sender verification is now available in
5016 $sender_address_data in subsequent conditions in the ACL statement.
5018 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
5020 15. Added a new option "connect=<time>" to callout options, to set a different
5023 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
5024 was its contents. (It was OK if the option was not defined at all.)
5026 17. A "Completed" log line is now written for messages that are removed from
5027 the spool by the -Mrm option.
5029 18. New variables $sender_verify_failure and $recipient_verify_failure contain
5030 information about exactly what failed.
5032 19. Added -dd to debug only the daemon process.
5034 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
5035 handles timeouts, both on the server side and network timeouts. Renamed the
5036 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
5038 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
5039 It is now set to "smtps".
5041 22. $host_address is now set to the target address during the checking of
5042 ignore_target_hosts.
5044 23. When checking ignore_target_hosts for an ipliteral router, no host name was
5045 being passed; this would have caused $sender_host_name to have been used if
5046 matching the list had actually called for a host name (not very likely,
5047 since this list is usually IP addresses). A host name is now passed as
5050 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5051 code that specifies a non-restarting handler (typically sigaction() in
5052 modern systems) in an attempt to fix a rare and obscure crash bug.
5054 25. Narrowed the window for a race in the daemon that could cause it to ignore
5055 SIGCHLD signals. This is not a major problem, because they are used only to
5056 wake it up if nothing else does.
5058 26. A malformed maildirsize file could cause Exim to calculate negative values
5059 for the mailbox size or file count. Odd effects could occur as a result.
5060 The maildirsize information is now recalculated if the size or filecount
5063 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5064 support for a long time. Removed HAVE_SYS_VFS_H.
5066 28. Installed the latest version of exipick from John Jetmore.
5068 29. In an address list, if the pattern was not a regular expression, an empty
5069 subject address (from a bounce message) matched only if the pattern was an
5070 empty string. Non-empty patterns were not even tested. This was the wrong
5071 because it is perfectly reasonable to use an empty address as part of a
5072 database query. An empty address is now tested by patterns that are
5073 lookups. However, all the other forms of pattern expect the subject to
5074 contain a local part and a domain, and therefore, for them, an empty
5075 address still always fails if the pattern is not itself empty.
5077 30. Exim went into a mad DNS loop when attempting to do a callout where the
5078 host was specified on an smtp transport, and looking it up yielded more
5079 than one IP address.
5081 31. Re-factored the code for checking spool and log partition space into a
5082 function that finds that data and another that does the check. The former
5083 is then used to implement four new variables: $spool_space, $log_space,
5084 $spool_inodes, and $log_inodes.
5086 32. The RFC2047 encoding function was originally intended for short strings
5087 such as real names; it was not keeping to the 75-character limit for
5088 encoded words that the RFC imposes. It now respects the limit, and
5089 generates multiple encoded words if necessary. To be on the safe side, I
5090 have increased the buffer size for the ${rfc2047: expansion operator from
5093 33. It is now permitted to omit both strings after an "if" condition; if the
5094 condition is true, the result is "true". As before, when the second string
5095 is omitted, a false condition yields an empty string. This makes it less
5096 cumbersome to write custom ACL and router conditions.
5098 34. Failure to deliver a bounce message always caused it to be frozen, even if
5099 there was an errors_to setting on the router. The errors_to setting is now
5102 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5103 canonical form (fully expanded) before being placed in
5104 $sender_host_address.
5106 36. The table in the code that translates DNS record types into text (T_A to
5107 "A" for instance) was missing entries for NS and CNAME. It is just possible
5108 that this could have caused confusion if both these types were looked up
5109 for the same domain, because the text type is used as part of Exim's
5110 per-process caching. But the chance of anyone hitting this buglet seems
5113 37. The dnsdb lookup has been extended in a number of ways.
5115 (1) There is a new type, "zns", which walks up the domain tree until it
5116 finds some nameserver records. It should be used with care.
5118 (2) There is a new type, "mxh", which is like "mx" except that it returns
5119 just the host names, not the priorities.
5121 (3) It is now possible to give a list of domains (or IP addresses) to be
5122 looked up. The behaviour when one of the lookups defers can be
5123 controlled by a keyword.
5125 (4) It is now possible to specify the separator character for use when
5126 multiple records are returned.
5128 38. The dnslists ACL condition has been extended: it is now possible to supply
5129 a list of IP addresses and/or domains to be looked up in a particular DNS
5132 39. Added log_selector=+queue_time_overall.
5134 40. When running the queue in the test harness, wait just a tad after forking a
5135 delivery process, to get repeatability of debugging output.
5137 41. Include certificate and key file names in error message when GnuTLS fails
5138 to set them up, because the GnuTLS error message doesn't include the name
5139 of the failing file when there is a problem reading it.
5141 42. Allow both -bf and -bF in the same test run.
5143 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
5145 44. The "Exiscan patch" is now merged into the mainline Exim source.
5147 45. Sometimes the final signoff response after QUIT could fail to get
5148 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5149 < 0 before doing a fflush(). This bug looks as though it goes back to the
5150 introduction of TLS in release 3.20, but "sometimes" must have been rare
5151 because the tests only now provoked it.
5153 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
5154 (this can affect the format of dates).
5156 47. exim_tidydb, when checking for the continued existence of a message for
5157 which it has found a message-specific retry record, was not finding
5158 messages that were in split spool directories. Consequently, it was
5159 deleting retry records that should have stayed in existence.
5161 48. Steve fixed some bugs in eximstats.
5163 49. The SPA authentication driver was not abandoning authentication and moving
5164 on to the next authenticator when an expansion was forced to fail,
5165 contradicting the general specification for all authenticators. Instead it
5166 was generating a temporary error. It now behaves as specified.
5168 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
5169 (the order specifies the preference for clients). The order is now AES256,
5170 AES128, 3DES, ARCFOUR128.
5172 51. Small patch to Sieve code - explicitly set From: when generating an
5175 52. Exim crashed if a remote delivery caused a very long error message to be
5176 recorded - for instance if somebody sent an entire SpamAssassin report back
5177 as a large number of 550 error lines. This bug was coincidentally fixed by
5178 increasing the size of one of Exim's internal buffers (big_buffer) that
5179 happened as part of the Exiscan merge. However, to be on the safe side, I
5180 have made the code more robust (and fixed the comments that describe what
5183 53. Now that there can be additional text after "Completed" in log lines (if
5184 the queue_time_overall log selector is set), a one-byte patch to exigrep
5185 was needed to allow it to recognize "Completed" as not the last thing in
5188 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
5189 patch that reportedly fixes this has been added. I am not expert enough to
5190 create a test for it. This is what the patch creator wrote:
5192 "I found a little strange behaviour of ldap code when working with
5193 Windows 2003 AD Domain, where users was placed in more than one
5194 Organization Units. When I tried to give exim partial DN, the exit code
5195 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
5196 But simultaneously result of request was absolutely normal ldap result,
5197 so I produce this patch..."
5199 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
5200 REFERENCE, so I have modified the code to exclude the patch when that macro
5203 55. Some experimental protocols are using DNS PTR records for new purposes. The
5204 keys for these records are domain names, not reversed IP addresses. The
5205 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5206 leaves it alone. Component reversal etc. now happens only for IP addresses.
5209 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
5211 57. Double the size of the debug message buffer (to 2048) so that more of very
5212 long debug lines gets shown.
5214 58. The exicyclog utility now does better if the number of log files to keep
5215 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
5217 59. Two changes related to the smtp_active_hostname option:
5219 (1) $smtp_active_hostname is now available as a variable.
5220 (2) The default for smtp_banner uses $smtp_active_hostname instead
5221 of $primary_hostname.
5223 60. The host_aton() function is supposed to be passed a string that is known
5224 to be a valid IP address. However, in the case of IPv6 addresses, it was
5225 not checking this. This is a hostage to fortune. Exim now panics and dies
5226 if the condition is not met. A case was found where this could be provoked
5227 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5228 components; fortuitously, this particular loophole had already been fixed
5229 by change 4.50/55 above.
5231 If there are any other similar loopholes, the new check in host_aton()
5232 itself should stop them being exploited. The report I received stated that
5233 data on the command line could provoke the exploit when Exim was running as
5234 exim, but did not say which command line option was involved. All I could
5235 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5236 running as the user.
5239 61. There was a buffer overflow vulnerability in the SPA authentication code
5240 (which came originally from the Samba project). I have added a test to the
5241 spa_base64_to_bits() function which I hope fixes it.
5244 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
5245 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
5247 63. The daemon start-up calls getloadavg() while still root for those OS that
5248 need the first call to be done as root, but it missed one case: when
5249 deliver_queue_load_max is set with deliver_drop_privilege. This is
5250 necessary for the benefit of the queue runner, because there is no re-exec
5251 when deliver_drop_privilege is set.
5253 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
5254 This has been fixed.
5256 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5257 was in use, was not putting the data itself into the right store pool;
5258 consequently, it could be overwritten for a subsequent message in the same
5259 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5262 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
5264 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
5265 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
5266 the function were treating the return as a boolean value, which happened to
5267 work because 0=false and not-0=true, but is not correct code.
5269 68. The host_aton() function was not handling scoped IPv6 addresses (those
5270 with, for example, "%eth0" on the end) correctly.
5272 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
5273 negated items (that is, ~something) in unsigned ints. Some compilers
5274 apparently mutter when there is no cast.
5276 70. If an address verification called from an ACL failed, and did not produce a
5277 user-specific message (i.e. there was only a "system" message), nothing was
5278 put in $acl_verify_message. In this situation, it now puts the system
5281 71. Change 4.23/11 added synchronization checking at the start of an SMTP
5282 session; change 4.31/43 added the unwanted input to the log line - except
5283 that it did not do this in the start of session case. It now does.
5285 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
5286 This is wrong and can cause the other end to generate a synchronization
5287 error if it is another Exim or anything else that does the synchronization
5288 check. A QUIT command is no longer sent after a timeout.
5290 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
5291 during host lookups.
5293 74. The defer_ok option of callout verification was not working if it was used
5294 when verifying addresses in header lines, that is, for this case:
5296 verify = header_sender/callout=defer_ok
5298 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
5299 those file descriptors could be used for SMTP connections. If anything
5300 wrote to stderr (the example that came up was "warn" in embedded Perl), it
5301 could be sent to the SMTP client, causing chaos. The daemon now opens
5302 stdin, stdout, and stderr to /dev/null when it puts itself into the
5305 76. Arrange for output from Perl's "warn" command to be written to Exim's main
5306 log by default. The user can override this with suitable Perl magic.
5308 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
5309 the log message when discard triggers, was not working for the DATA ACL or
5310 for the non-SMTP ACL.
5312 78. Error message wording change in sieve.c.
5314 79. If smtp_accept_max_per_host was set, the number of connections could be
5315 restricted to fewer than expected, because the daemon was trying to set up
5316 a new connection before checking whether the processes handling previous
5317 connections had finished. The check for completed processes is now done
5318 earlier. On busy systems, this bug wouldn't be noticed because something
5319 else would have woken the daemon, and it would have reaped the completed
5322 80. If a message was submitted locally by a user whose login name contained one
5323 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
5324 It caused a spool format error. I have fixed the spool reading code. A
5325 related problem was that the "from" clause in the Received: line became
5326 illegal because of the space(s). It is now covered by ${quote_local_part.
5328 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
5331 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
5333 83. Patch from Sieve maintainer to fix unterminated string problem in
5334 "vacation" handling.
5336 84. Some minor changes to the Linux configuration files to help with other
5337 OS variants using glibc.
5339 85. One more patch for Sieve to update vacation handling to latest spec.
5342 ----------------------------------------------------
5343 See the note above about the 4.44 and 4.50 releases.
5344 ----------------------------------------------------
5350 1. Change 4.43/35 introduced a bug that caused file counts to be
5351 incorrectly computed when quota_filecount was set in an appendfile
5354 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
5355 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
5358 3. Give more explanation in the error message when the command for a transport
5359 filter fails to execute.
5361 4. There are several places where Exim runs a non-Exim command in a
5362 subprocess. The SIGUSR1 signal should be disabled for these processes. This
5363 was being done only for the command run by the queryprogram router. It is
5364 now done for all such subprocesses. The other cases are: ${run, transport
5365 filters, and the commands run by the lmtp and pipe transports.
5367 5. Some older OS have a limit of 256 on the maximum number of file
5368 descriptors. Exim was using setrlimit() to set 1000 as a large value
5369 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
5370 systems. I've change it so that if it can't get 1000, it tries for 256.
5372 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
5373 was an oversight, and furthermore, ever since the addition of extra
5374 controls (e.g. 4.43/32), the checks on when to allow different forms of
5375 "control" were broken. There should now be diagnostics for all cases when a
5376 control that does not make sense is encountered.
5378 7. $recipients is now available in the predata ACL (oversight).
5380 8. Tidy the search cache before the fork to do a delivery from a message
5381 received from the command line. Otherwise the child will trigger a lookup
5382 failure and thereby defer the delivery if it tries to use (for example) a
5383 cached ldap connection that the parent has called unbind on.
5385 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
5386 of $address_data from the recipient verification was clobbered by the
5387 sender verification.
5389 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
5390 was its contents. (It was OK if the option was not defined at all.)
5392 11. A "Completed" log line is now written for messages that are removed from
5393 the spool by the -Mrm option.
5395 12. $host_address is now set to the target address during the checking of
5396 ignore_target_hosts.
5398 13. When checking ignore_target_hosts for an ipliteral router, no host name was
5399 being passed; this would have caused $sender_host_name to have been used if
5400 matching the list had actually called for a host name (not very likely,
5401 since this list is usually IP addresses). A host name is now passed as
5404 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5405 code that specifies a non-restarting handler (typically sigaction() in
5406 modern systems) in an attempt to fix a rare and obscure crash bug.
5408 15. Narrowed the window for a race in the daemon that could cause it to ignore
5409 SIGCHLD signals. This is not a major problem, because they are used only to
5410 wake it up if nothing else does.
5412 16. A malformed maildirsize file could cause Exim to calculate negative values
5413 for the mailbox size or file count. Odd effects could occur as a result.
5414 The maildirsize information is now recalculated if the size or filecount
5417 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5418 support for a long time. Removed HAVE_SYS_VFS_H.
5420 18. Updated exipick to current release from John Jetmore.
5422 19. Allow an empty sender to be matched against a lookup in an address list.
5423 Previously the only cases considered were a regular expression, or an
5426 20. Exim went into a mad DNS lookup loop when doing a callout where the
5427 host was specified on the transport, if the DNS lookup yielded more than
5430 21. The RFC2047 encoding function was originally intended for short strings
5431 such as real names; it was not keeping to the 75-character limit for
5432 encoded words that the RFC imposes. It now respects the limit, and
5433 generates multiple encoded words if necessary. To be on the safe side, I
5434 have increased the buffer size for the ${rfc2047: expansion operator from
5437 22. Failure to deliver a bounce message always caused it to be frozen, even if
5438 there was an errors_to setting on the router. The errors_to setting is now
5441 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5442 canonical form (fully expanded) before being placed in
5443 $sender_host_address.
5445 24. Updated eximstats to version 1.33
5447 25. Include certificate and key file names in error message when GnuTLS fails
5448 to set them up, because the GnuTLS error message doesn't include the name
5449 of the failing file when there is a problem reading it.
5451 26. Expand error message when OpenSSL has problems setting up cert/key files.
5454 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
5455 (this can affect the format of dates).
5457 28. exim_tidydb, when checking for the continued existence of a message for
5458 which it has found a message-specific retry record, was not finding
5459 messages that were in split spool directories. Consequently, it was
5460 deleting retry records that should have stayed in existence.
5462 29. eximstats updated to version 1.35
5463 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
5464 1.35 - bugfix such that pie charts by volume are generated correctly
5466 30. The SPA authentication driver was not abandoning authentication and moving
5467 on to the next authenticator when an expansion was forced to fail,
5468 contradicting the general specification for all authenticators. Instead it
5469 was generating a temporary error. It now behaves as specified.
5471 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
5472 (the order specifies the preference for clients). The order is now AES256,
5473 AES128, 3DES, ARCFOUR128.
5475 31. Small patch to Sieve code - explicitly set From: when generating an
5478 32. Exim crashed if a remote delivery caused a very long error message to be
5479 recorded - for instance if somebody sent an entire SpamAssassin report back
5480 as a large number of 550 error lines. This bug was coincidentally fixed by
5481 increasing the size of one of Exim's internal buffers (big_buffer) that
5482 happened as part of the Exiscan merge. However, to be on the safe side, I
5483 have made the code more robust (and fixed the comments that describe what
5486 33. Some experimental protocols are using DNS PTR records for new purposes. The
5487 keys for these records are domain names, not reversed IP addresses. The
5488 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5489 leaves it alone. Component reversal etc. now happens only for IP addresses.
5492 34. The host_aton() function is supposed to be passed a string that is known
5493 to be a valid IP address. However, in the case of IPv6 addresses, it was
5494 not checking this. This is a hostage to fortune. Exim now panics and dies
5495 if the condition is not met. A case was found where this could be provoked
5496 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5497 components; fortuitously, this particular loophole had already been fixed
5498 by change 4.50/55 or 4.44/33 above.
5500 If there are any other similar loopholes, the new check in host_aton()
5501 itself should stop them being exploited. The report I received stated that
5502 data on the command line could provoke the exploit when Exim was running as
5503 exim, but did not say which command line option was involved. All I could
5504 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5505 running as the user.
5508 35. There was a buffer overflow vulnerability in the SPA authentication code
5509 (which came originally from the Samba project). I have added a test to the
5510 spa_base64_to_bits() function which I hope fixes it.
5513 36. The daemon start-up calls getloadavg() while still root for those OS that
5514 need the first call to be done as root, but it missed one case: when
5515 deliver_queue_load_max is set with deliver_drop_privilege. This is
5516 necessary for the benefit of the queue runner, because there is no re-exec
5517 when deliver_drop_privilege is set.
5519 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5520 was in use, was not putting the data itself into the right store pool;
5521 consequently, it could be overwritten for a subsequent message in the same
5522 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5525 38. Sometimes the final signoff response after QUIT could fail to get
5526 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5527 < 0 before doing a fflush(). This bug looks as though it goes back to the
5528 introduction of TLS in release 3.20, but "sometimes" must have been rare
5529 because the tests only now provoked it.
5535 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5536 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5537 Now it returns an integer. A number of calls were still expecting a T/F
5538 return. Fortuitously, in all cases, the tests worked in OK situations,
5539 which is the norm. However, things would have gone wrong on any write
5540 failures on the smtp file descriptor. This function is used when sending
5541 messages over SMTP and also when doing verify callouts.
5543 2. When Exim is called to do synchronous delivery of a locally submitted
5544 message (the -odf or -odi options), it no longer closes stderr before doing
5547 3. Implemented the mua_wrapper option.
5549 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5551 5. Implemented the functions header_remove(), header_testname(),
5552 header_add_at_position(), and receive_remove_recipient(), and exported them
5555 6. If an ACL "warn" statement specified the addition of headers, Exim already
5556 inserted X-ACL-Warn: at the start if there was no header name. However, it
5557 was not making this test for the second and subsequent header lines if
5558 there were newlines in the string. This meant that an invalid header could
5559 be inserted if Exim was badly configured.
5561 7. Allow an ACL "warn" statement to add header lines at the start or after all
5562 the Received: headers, as well as at the end.
5564 8. Added the rcpt_4xx retry error code.
5566 9. Added postmaster_mailfrom=xxx to callout verification option.
5568 10. Added mailfrom=xxxx to the callout verification option, for verify=
5571 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5572 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5574 12. Inserted some casts to stop certain compilers warning when using pointer
5575 differences as field lengths or precisions in printf-type calls (mostly
5576 affecting debugging statements).
5578 13. Added optional readline() support for -be (dynamically loaded).
5580 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5581 same clock tick as a message's arrival, so that its received time was the
5582 same as the "first fail" time on the retry record, and that message
5583 remained on the queue past the ultimate address timeout, every queue runner
5584 would try a delivery (because it was past the ultimate address timeout) but
5585 after another failure, the ultimate address timeout, which should have then
5586 bounced the address, did not kick in. This was a "< instead of <=" error;
5587 in most cases the first failure would have been in the next clock tick
5588 after the received time, and all would be well.
5590 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5591 being recognized when the domain list was tested by the match_domain
5592 condition in an expansion string.
5594 16. Added the ${str2b64: operator.
5596 17. Exim was always calling setrlimit() to set a large limit for the number of
5597 processes, without checking whether the existing limit was already
5598 adequate. (It did check for the limit on file descriptors.) Furthermore,
5599 errors from getrlimit() and setrlimit() were being ignored. Now they are
5600 logged to the main and panic logs, but Exim does carry on, to try to do its
5601 job under whatever limits there are.
5603 18. Imported PCRE 5.0.
5605 19. Trivial typo in log message " temporarily refused connection" (the leading
5608 20. If the log selector return_path_on_delivery was set and an address was
5609 redirected to /dev/null, the delivery process crashed because it assumed
5610 that a return path would always be set for a "successful" delivery. In this
5611 case, the whole delivery is bypassed as an optimization, and therefore no
5614 21. Internal re-arrangement: the function for sending a challenge and reading
5615 a response while authentication was assuming a zero-terminated challenge
5616 string. It's now changed to take a pointer and a length, to allow for
5617 binary data in such strings.
5619 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5621 23. Exim was not respecting finduser_retries when seeking the login of the
5622 uid under which it was called; it was always trying 10 times. (The default
5623 setting of finduser_retries is zero.) Also, it was sleeping after the final
5624 failure, which is pointless.
5626 24. Implemented tls_on_connect_ports.
5628 25. Implemented acl_smtp_predata.
5630 26. If the domain in control=submission is set empty, Exim assumes that the
5631 authenticated id is a complete email address when it generates From: or
5632 Sender: header lines.
5634 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5635 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5636 chown and chgrp in /bin and hostname in /usr/bin.
5638 28. Exim was keeping the "process log" file open after each use, just as it
5639 does for the main log. This opens the possibility of it remaining open for
5640 long periods when the USR1 signal hits a daemon. Occasional processlog
5641 errors were reported, that could have been caused by this. Anyway, it seems
5642 much more sensible not to leave this file open at all, so that is what now
5645 29. The long-running daemon process does not normally write to the log once it
5646 has entered its main loop, and it closes the log before doing so. This is
5647 so that log files can straightforwardly be renamed and moved. However,
5648 there are a couple of unusual error situations where the daemon does write
5649 log entries, and I had neglected to close the log afterwards.
5651 30. The text of an SMTP error response that was received during a remote
5652 delivery was being truncated at 512 bytes. This is too short for some of
5653 the long messages that one sometimes sees. I've increased the limit to
5656 31. It is now possible to make retry rules that apply only when a message has a
5657 specific sender, in particular, an empty sender.
5659 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5660 it possible to be selective about when SMTP synchronization is enforced.
5662 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5664 32. Implemented hosts_connection_nolog.
5666 33. Added an ACL for QUIT.
5668 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5671 35. Added mailbox_size and mailbox_filecount to appendfile.
5673 36. Added control = no_multiline_responses to ACLs.
5675 37. There was a bug in the logic of the code that waits for the clock to tick
5676 in the case where the clock went backwards by a substantial amount such
5677 that the microsecond fraction of "now" was more than the microsecond
5678 fraction of "then" (but the whole seconds number was less).
5680 38. Added support for the libradius Radius client library this is found on
5681 FreeBSD (previously only the radiusclient library was supported).
5687 1. When certain lookups returned multiple values in the form name=value, the
5688 quoting of the values was not always being done properly. Specifically:
5689 (a) If the value started with a double quote, but contained no whitespace,
5691 (b) If the value contained whitespace other than a space character (i.e.
5692 tabs or newlines or carriage returns) it was not quoted.
5693 This fix has been applied to the mysql and pgsql lookups by writing a
5694 separate quoting function and calling it from the lookup code. The fix
5695 should probably also be applied to nisplus, ibase and oracle lookups, but
5696 since I cannot test any of those, I have not disturbed their existing code.
5698 2. A hit in the callout cache for a specific address caused a log line with no
5699 reason for rejecting RCPT. Now it says "Previous (cached) callout
5700 verification failure".
5702 3. There was an off-by-one bug in the queryprogram router. An over-long
5703 return line was truncated at 256 instead of 255 characters, thereby
5704 overflowing its buffer with the terminating zero. As well as fixing this, I
5705 have increased the buffer size to 1024 (and made a note to document this).
5707 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5708 when Exim is waiting for an SMTP response from a remote server, Exim
5709 restarts its select() call on the socket, thereby resetting its timeout.
5710 This is not a problem when such interrupts are rare. Somebody set up a cron
5711 job to run exiwhat every 2 minutes, which is less than the normal select()
5712 timeout (5 or 10 minutes). This meant that the select() timeout never
5713 kicked in because it was always reset. I have fixed this by comparing the
5714 time when an interrupt arrives with the time at the start of the first call
5715 to select(). If more time than the timeout has elapsed, the interrupt is
5716 treated as a timeout.
5718 5. Some internal re-factoring in preparation for the addition of Sieve
5719 extensions (by MH). In particular, the "personal" test is moved to a
5720 separate function, and given an option for scanning Cc: and Bcc: (which is
5721 not set for Exim filters).
5723 6. When Exim created an email address using the login of the caller as the
5724 local part (e.g. when creating a From: or Sender: header line), it was not
5725 quoting the local part when it contained special characters such as @.
5727 7. Installed new OpenBSD configuration files.
5729 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5730 try to make them clearer.
5732 9. Callout options, other than the timeout value, were being ignored when
5733 verifying sender addresses in header lines. For example, when using
5735 verify = header_sender/callout=no_cache
5737 the cache was (incorrectly) being used.
5739 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5740 only the Cygwin environment.
5742 11. When return_path_on_delivery was set as a log selector, if different remote
5743 addresses in the same message used different return paths and parallel
5744 remote delivery occurred, the wrong values would sometimes be logged.
5745 (Whenever a remote delivery process finished, the return path value from
5746 the most recently started remote delivery process was logged.)
5748 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5749 lines when AUTH and/or TLS are in use. This is the "received protocol"
5750 field. Exim used to use "asmtp" for authenticated SMTP, without any
5751 indication (in the protocol name) for TLS use. Now it follows the RFC and
5752 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5753 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5754 names appear in log lines as well as in Received: header lines.
5756 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5757 extensions, and comparison tests, and to fix some bugs.
5759 14. Changes to the "personal" filter test:
5761 (1) The test was buggy in that it was just doing the equivalent of
5762 "contains" tests on header lines. For example, if a user's address was
5763 anne@some.where, the "personal" test would incorrectly be true for
5765 To: susanne@some.where
5767 This test is now done by extracting each address from the header in turn,
5768 and checking the entire address. Other tests that are part of "personal"
5769 are now done using regular expressions (for example, to check local parts
5770 of addresses in From: header lines).
5772 (2) The list of non-personal local parts in From: addresses has been
5773 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5774 taken from the Sieve specification recommendations.
5776 (3) If the message contains any header line starting with "List-" it is
5777 treated as non-personal.
5779 (4) The test for "circular" in the Subject: header line has been removed
5780 because it now seems ill-conceived.
5782 15. Minor typos in src/EDITME comments corrected.
5784 16. Installed latest exipick from John Jetmore.
5786 17. If headers_add on a router specified a text string that was too long for
5787 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5788 of string_sprintf() is now avoided.
5790 18. $message_body_size was not set (it was always zero) when running the DATA
5791 ACL and the local_scan() function.
5793 19. For the "mail" command in an Exim filter, no default was being set for
5794 the once_repeat time, causing a random time value to be used if "once" was
5795 specified. (If the value happened to be <= 0, no repeat happened.) The
5796 default is now 0s, meaning "never repeat". The "vacation" command was OK
5797 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5798 (I found it when inspecting the code).
5800 20. There is now an overall timeout for performing a callout verification. It
5801 defaults to 4 times the callout timeout, which applies to individual SMTP
5802 commands during the callout. The overall timeout applies when there is more
5803 than one host that can be tried. The timeout is checked before trying the
5804 next host. This prevents very long delays if there are a large number of
5805 hosts and all are timing out (e.g. when the network connections are timing
5806 out). The value of the overall timeout can be changed by specifying an
5807 additional sub-option for "callout", called "maxwait". For example:
5809 verify = sender/callout=5s,maxwait=20s
5811 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5812 to the end before writing, but this should make it even safer).
5814 22. Exim was forgetting that it had advertised PIPELINING for the second and
5815 subsequent messages on an SMTP connection. It was also not resetting its
5816 memory on STARTTLS and an internal HELO.
5818 23. When Exim logs an SMTP synchronization error within a session, it now
5819 records whether PIPELINING has been advertised or not.
5821 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5822 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5823 rather than long int.
5825 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5827 26. Added the never_mail option to autoreply.
5833 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5834 crash if the getsockname() call failed; this can happen if a connection is
5835 closed very soon after it is established. The problem was simply in the
5836 order in which certain operations were done, causing Exim to try to write
5837 to the SMTP stream before it had set up the file descriptor. The bug has
5838 been fixed by making things happen in the correct order.
5844 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5845 before the connection was closed, thus losing the rejection response.
5847 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5848 some early Solaris releases, but causes trouble in current releases where
5849 socklen_t is defined.
5851 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5854 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5857 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5858 that releasing the top of it at the end releases what was used for sub-
5859 expansions (unless the block got too big). However, discard this block if
5860 the first thing is a variable or header, so that we can use its block when
5861 it is dynamic (useful for very large $message_headers, for example).
5863 6. Lookups now cache *every* query, not just the most recent. A new, separate
5864 store pool is used for this. It can be recovered when all lookup caches are
5865 flushed. Lookups now release memory at the end of their result strings.
5866 This has involved some general refactoring of the lookup sources.
5868 7. Some code has been added to the store_xxx() functions to reduce the amount
5869 of flapping under certain conditions.
5871 8. log_incoming_interface used to affect only the <= reception log lines. Now
5872 it causes the local interface and port to be added to several more SMTP log
5873 lines, for example "SMTP connection from", and rejection lines.
5875 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5877 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5879 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5880 could be overwritten at the end of the current message (or the start of a
5881 new message if it was set in a HELO ACL). The value is now preserved for
5882 the duration of the SMTP connection.
5884 12. If a transport had a headers_rewrite setting, and a matching header line
5885 contained an unqualified address, that address was qualified, even if it
5886 did not match any rewriting rules. The underlying bug was that the values
5887 of the flags that permit the existence of unqualified sender and recipient
5888 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5889 non-local messages, and by -bnq for local messages) were not being
5890 preserved with the message after it was received.
5892 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5893 "next input=" as part of the text comprising the host identity instead of
5894 the correct text. The code was using the same buffer for two different
5895 strings. However, depending on which order the printing function evaluated
5896 its arguments, the bug did not always show up. Under Linux, for example, my
5897 test suite worked just fine.
5899 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5900 doesn't work with some older versions of Perl. It has been changed to "my",
5901 which in any case is probably the better facility to use.
5903 15. A really picky compiler found some instances of statements for creating
5904 error messages that either had too many or two few arguments for the format
5907 16. The size of the buffer for calls to the DNS resolver has been increased
5908 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5909 for addresses that have a lot of PTR records. This alleviates a problem; it
5910 does not fully solve it.
5912 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5913 buffer now truncates the list and logs the incident, which is the same
5914 action as happens when Exim is looking up a host name and its aliases.
5915 Previously in this situation something unpredictable would happen;
5916 sometimes it was "internal error: store_reset failed".
5918 18. If a server dropped the connection unexpectedly when an Exim client was
5919 using GnuTLS and trying to read a response, the client delivery process
5920 crashed while trying to generate an error log message.
5922 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5923 string, for example:
5925 warn message = H1: something\nH2: something
5927 the text was added as a single header line from Exim's point of view
5928 though it ended up OK in the delivered message. However, searching for the
5929 second and subsequent header lines using $h_h2: did not work. This has been
5930 fixed. Similarly, if a system filter added multiple headers in this way,
5931 the routers could not see them.
5933 20. Expanded the error message when iplsearch is called with an invalid key to
5934 suggest using net-iplsearch in a host list.
5936 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5937 ACLs are no longer actually imposed (and a message to that effect is
5940 22. If a "gecos" field in a passwd entry contained escaped characters, in
5941 particular, if it contained a \" sequence, Exim got it wrong when building
5942 a From: or a Sender: header from that name. A second bug also caused
5943 incorrect handling when an unquoted " was present following a character
5944 that needed quoting.
5946 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5947 was not being matched caselessly.
5949 24. Arranged for all hyphens in the exim.8 source to be escaped with
5952 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5953 properly. Recipient callout cache records were still being keyed to include
5954 the sender, even when use_sender was set false. This led to far more
5955 callouts that were necessary. The sender is no longer included in the key
5956 when use_sender is false.
5958 26. Added "control = submission" modifier to ACLs.
5960 27. Added the ${base62d: operator to decode base 62 numbers.
5962 28. dnsdb lookups can now access SRV records.
5964 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5965 the configuration file.
5967 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5968 (-v) mode. This makes the output for a verbose queue run more intelligible.
5970 31. Added a use_postmaster feature to recipient callouts.
5972 32. Added the $body_zerocount variable, containing the number of binary zero
5973 bytes in the message body.
5975 33. The time of last modification of the "new" subdirectory is now used as the
5976 "mailbox time last read" when there is a quota error for a maildir
5979 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5981 35. Added +ignore_unknown as a special item in host lists.
5983 36. Code for decoding IPv6 addresses in host lists is now included, even if
5984 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5985 address was recognized as an IP address, but was then not correctly decoded
5986 into binary, causing unexpected and incorrect effects when compared with
5993 1. Very minor rewording of debugging text in manualroute to say "list of
5994 hosts" instead of "hostlist".
5996 2. If verify=header_syntax was set, and a header line with an unqualified
5997 address (no domain) and a large number of spaces between the end of the
5998 name and the colon was received, the reception process suffered a buffer
5999 overflow, and (when I tested it) crashed. This was caused by some obsolete
6000 code that should have been removed. The fix is to remove it!
6002 3. When running in the test harness, delay a bit after writing a bounce
6003 message to get a bit more predictability in the log output.
6005 4. Added a call to search_tidyup() just before forking a reception process. In
6006 theory, someone could use a lookup in the expansion of smtp_accept_max_
6007 per_host which, without the tidyup, could leave open a database connection.
6009 5. Added the variables $recipient_data and $sender_data which get set from a
6010 lookup success in an ACL "recipients" or "senders" condition, or a router
6011 "senders" option, similar to $domain_data and $local_part_data.
6013 6. Moved the writing of debug_print from before to after the "senders" test
6016 7. Change 4.31/66 (moving the time when the Received: is generated) caused
6017 problems for message scanning, either using a data ACL, or using
6018 local_scan() because the Received: header was not generated till after they
6019 were called (in order to set the time as the time of reception completion).
6020 I have revised the way this works. The header is now generated after the
6021 body is received, but before the ACL or local_scan() are called. After they
6022 are run, the timestamp in the header is updated.
6028 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
6029 before starting a queue runner without re-exec. This happened only when
6030 deliver_drop_privilege was set or when the Exim user was set to root. The
6031 effect of the bug was that timeouts during subsequent deliveries caused
6032 crashes instead of being properly handled. The handler is now left at its
6033 default (and expected) setting.
6035 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
6036 message, again when deliver_drop_privilege is set or Exim is run as root.
6037 The bug described in (1) was not present in this case, but the tidying up
6038 of the other signals was missing. I have made the two cases consistent.
6040 3. The ignore_target_hosts setting on a manualroute router was being ignored
6041 for hosts that were looked up using the /MX notation.
6043 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
6046 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
6047 operated on the sender address. After changing the $sender_address to <>
6048 for the sender address verify, Exim was re-instated it as the original
6049 (before rewriting) address, but remembering that it had rewritten it, so it
6050 wasn't rewriting it again. This bug also had the effect of breaking the
6051 sender address verification caching when the sender address was rewritten.
6053 6. The ignore_target_hosts option was being ignored by the ipliteral router.
6054 This has been changed so that if the ip literal address matches
6055 ignore_target_hosts, the router declines.
6057 7. Added expansion conditions match_domain, match_address, and match_local_
6058 part (NOT match_host).
6060 8. The placeholder for the Received: header didn't have a length field set.
6062 9. Added code to Exim itself and to exim_lock to test for a specific race
6063 condition that could lead to file corruption when using MBX delivery. The
6064 issue is with the lockfile that is created in /tmp. If this file is removed
6065 after a process has opened it but before that process has acquired a lock,
6066 there is the potential for a second process to recreate the file and also
6067 acquire a lock. This could lead to two Exim processes writing to the file
6068 at the same time. The added code performs the same test as UW imapd; it
6069 checks after acquiring the lock that its file descriptor still refers to
6070 the same named file.
6072 10. The buffer for building added header lines was of fixed size, 8192 bytes.
6073 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
6076 11. Added the smtp_active_hostname option. If used, this will typically be made
6077 to depend on the incoming interface address. Because $interface_address is
6078 not set up until the daemon has forked a reception process, error responses
6079 that can happen earlier (such as "too many connections") no longer contain
6082 12. If an expansion in a condition on a "warn" statement fails because a lookup
6083 defers, the "warn" statement is abandoned, and the next ACL statement is
6084 processed. Previously this caused the whole ACL to be aborted.
6086 13. Added the iplsearch lookup type.
6088 14. Added ident_timeout as a log selector.
6090 15. Added tls_certificate_verified as a log selector.
6092 16. Added a global option tls_require_ciphers (compare the smtp transport
6093 option of the same name). This controls incoming TLS connections.
6095 17. I finally figured out how to make tls_require_ciphers do a similar thing
6096 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
6097 before starting the TLS session.
6099 18. Tabs are now shown as \t in -bP output.
6101 19. If the log selector return_path_on_delivery was set, Exim crashed when
6102 bouncing a message because it had too many Received: header lines.
6104 20. If two routers both had headers_remove settings, and the first one included
6105 a superfluous trailing colon, the final name in the first list and the
6106 first name in the second list were incorrectly joined into one item (with a
6107 colon in the middle).
6113 1. Added -C and -D options to the exinext utility, mainly to make it easier
6114 to include in the automated testing, but these could be helpful when
6115 multiple configurations are in use.
6117 2. The exinext utility was not formatting the output nicely when there was
6118 an alternate port involved in the retry record key, nor when there was a
6119 message id as well (for retries that were specific to a specific message
6120 and a specific host). It was also confused by IPv6 addresses, because of
6121 the additional colons they contain. I have fixed the IPv4 problem, and
6122 patched it up to do a reasonable job for IPv6.
6124 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
6125 delivery, the log line now contains "pipelined" if PIPELINING was used.
6127 4. An SMTP transport process used to panic and die if the bind() call to set
6128 an explicit outgoing interface failed. This has been changed; it is now
6129 treated in the same way as a connect() failure.
6131 5. A reference to $sender_host_name in the part of a conditional expansion
6132 that was being skipped was still causing a DNS lookup. This no longer
6135 6. The def: expansion condition was not recognizing references to header lines
6136 that used bh_ and bheader_.
6138 7. Added the _cache feature to named lists.
6140 8. The code for checking quota_filecount in the appendfile transport was
6141 allowing one more file than it should have been.
6143 9. For compatibility with Sendmail, the command line option
6151 and sets the incoming protocol and host name (for trusted callers). The
6152 host name and its colon can be omitted when only the protocol is to be set.
6153 Note the Exim already has two private options, -pd and -ps, that refer to
6154 embedded Perl. It is therefore impossible to set a protocol value of "d" or
6155 "s", but I don't think that's a major issue.
6157 10. A number of refactoring changes to the code, none of which should affect
6160 (a) The number of logging options was getting close to filling up the
6161 32-bit word that was used as a bit map. I have split them into two classes:
6162 those that are passed in the argument to log_write(), and those that are
6163 only ever tested independently outside of that function. These are now in
6164 separate 32-bit words, so there is plenty of room for expansion again.
6165 There is no change in the user interface or the logging behaviour.
6167 (b) When building, for example, log lines, the code previously used a
6168 macro that called string_cat() twice, in order to add two strings. This is
6169 not really sufficiently general. Furthermore, there was one instance where
6170 it was actually wrong because one of the argument was used twice, and in
6171 one call a function was used. (As it happened, calling the function twice
6172 did not affect the overall behaviour.) The macro has been replaced by a
6173 function that can join an arbitrary number of extra strings onto a growing
6176 (c) The code for expansion conditions now uses a table and a binary chop
6177 instead of a serial search (which was left over from when there were very
6178 few conditions). Also, it now recognizes conditions like "pam" even when
6179 the relevant support is not compiled in: a suitably worded error message is
6180 given if an attempt is made to use such a condition.
6182 11. Added ${time_interval:xxxxx}.
6184 12. A bug was causing one of the ddress fields not to be passed back correctly
6185 from remote delivery subprocesses. The field in question was not being
6186 subsequently used, so this caused to problems in practice.
6188 13. Added new log selectors queue_time and deliver_time.
6190 14. Might have fixed a bug in maildirsizefile handling that threw up
6191 "unexpected character" debug warnings, and recalculated the data
6192 unnecessarily. In any case, I expanded the warning message to give more
6195 15. Added the message "Restricted characters in address" to the statements in
6196 the default ACL that block characters like @ and % in local parts.
6198 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
6199 Three changes have been made:
6201 (a) There was a serious bug; a negative response to MAIL caused the whole
6202 recipient domain to be cached as invalid, thereby blocking all messages
6203 to all local parts at the same domain, from all senders. This bug has
6204 been fixed. The domain is no longer cached after a negative response to
6205 MAIL if the sender used is not empty.
6207 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
6210 (c) A new callout option, "use_sender" has been added for people who want
6211 the modified behaviour.
6217 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
6220 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
6221 indeed breaks things for older releases.
6223 3. Added additional logging to the case where there is a problem reading data
6224 from a filter that is running in a subprocess using a pipe, in order to
6225 try to track down a specific problem.
6227 4. Testing facility fudge: when running in the test harness and attempting
6228 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
6229 getting "No route to host". Convert this to a timeout.
6231 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
6234 6. Some OS don't have socklen_t but use size_t instead. This affects the
6235 fifth argument of getsockopt() amongst other things. This is now
6236 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
6237 can be set for individual OS. I have set it for SunOS5, OSF1, and
6238 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
6239 some earlier ones do not.
6241 7. Change 4.30/15 was not doing the test caselessly.
6243 8. The standard form for an IPv6 address literal was being rejected by address
6244 parsing in, for example, MAIL and RCPT commands. An example of this kind of
6245 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6246 this, as well as the form without the "IPv6" on the front (but only when
6247 address literals are enabled, of course).
6249 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
6251 10. Exim crashed if a message with an empty sender address specified by -f
6252 encountered a router with an errors_to setting. This could be provoked only
6253 by a command such as
6257 where an empty string was supplied; "<>" did not hit this bug.
6259 11. Installed PCRE release 4.5.
6261 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
6262 remained set. It is now erased.
6264 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
6265 times from message ids (which are base 36 rather than the normal 62).
6267 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
6268 were being counted as actual protocol errors, and logged if the log
6269 selector +smtp_protocol_error was set. One cannot be perfect in this test,
6270 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
6271 and DATA following a set of rejected RCPTs do not count as protocol errors.
6272 In other words, Exim assumes they were pipelined, though this may not
6273 actually be the case. Of course, in all cases the client gets an
6274 appropriate error code.
6276 15. If a lookup fails in an ACL condition, a message about the failure may
6277 be available; it is used if testing the ACL cannot continue, because most
6278 such messages specify what the cause of the deferral is. However, some
6279 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
6280 that caused an old message to be retained and used if a later statement
6281 caused a defer, replacing the real cause of the deferral.
6283 16. If an IP address had so many PTR records that the DNS lookup buffer
6284 was not large enough to hold them, Exim could crash while trying to process
6285 the truncated data. It now detects and logs this case.
6287 17. Further to 4.21/58, another change has been made: if (and only if) the
6288 first line of a message (the first header line) ends with CRLF, a bare LF
6289 in a subsequent header line has a space inserted after it, so as not to
6290 terminate the header.
6292 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
6293 unnecessarily, used atoi() instead of strtol(), and didn't check the
6294 termination when getting file sizes from file names by regex.
6296 19. Completely re-implemented the support for maildirsize files, in the light
6297 of a number of problems with the previous contributed implementation
6298 (4.30/29). In particular:
6300 . If the quota is zero, the maildirsize file is maintained, but no quota is
6303 . If the maildir directory does not exist, it is created before any attempt
6304 to write a maildirsize file.
6306 . The quota value in the file is just a cache; if the quota is changed in
6307 the transport, the new value overrides.
6309 . A regular expression is available for excluding directories from the
6312 20. The autoreply transport checks the characters in options that define the
6313 message's headers; it allows continued headers, but it was checking with
6314 isspace() after an embedded newline instead of explicitly looking for a
6317 21. If all the "regular" hosts to which an address was routed had passed their
6318 expiry times, and had not reached their retry times, the address was
6319 bounced, even if fallback hosts were defined. Now Exim should go on to try
6322 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
6323 equivalent code in the SMTP transport. Some hosts send humungous responses
6324 to HELO/EHLO, more than 1024 it seems.
6326 23. Refactoring: code in filter.c used (void *) for "any old type" but this
6327 gives compiler warnings in some environments. I've now done it "properly",
6330 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
6331 (because of problems with the built-in one) was declared to return uschar *
6332 instead of char *, causing compiler failure.
6334 25. Fixed a file descriptor leak when processing alias/forward files.
6336 26. Fixed a minor format string issue in dbfn.c.
6338 27. Typo in exim.c: ("dmbnz" for "dbmnz").
6340 28. If a filter file refered to $h_xxx or $message_headers, and the headers
6341 contained RFC 2047 "words", Exim's memory could, under certain conditions,
6344 29. When a sender address is verified, it is cached, to save repeating the test
6345 when there is more than one recipient in a message. However, when the
6346 verification involves a callout, it is possible for different callout
6347 options to be set for different recipients. It is too complicated to keep
6348 track of this in the cache, so now Exim always runs a verification when a
6349 callout is required, relying on the callout cache for the optimization.
6350 The overhead is duplication of the address routing, but this should not be
6353 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
6354 to be verified with callout=postmaster, and the main callout worked but the
6355 postmaster check failed, the verification correctly failed. However, if a
6356 subsequent RCPT command asked for sender verification *without* the
6357 postmaster check, incorrect caching caused this verification also to fail,
6360 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
6361 it was not caching the DNS options (qualify_single, search_parents) that
6362 were used when the lookup failed. A subsequent lookup with different
6363 options therefore always gave the same answer, though there were cases
6364 where it should not have. (Example: a "domains = !$mx_any" option on a
6365 dnslookup router: the "domains" option is always processed without any
6366 widening, but the router might have qualify_single set.) Now Exim uses the
6367 cached value only when the same options are set.
6369 32. Added John Jetmore's "exipick" utility to the distribution.
6371 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
6372 than a timeout (e.g. a certificate is required, and is not provided), an
6373 Exim server now closes the connection immediately. Previously it waited for
6374 the client to close - but if the client is SSL, it seems that they each
6375 wait for each other, leading to a delay before one of them times out.
6377 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
6378 maintained 0.8.x compatibility because I don't think many are using it, and
6379 it is clearly obsolete.
6381 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
6384 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
6385 client certificate was expired. A simple patch fixes this, though I don't
6386 understand the full logic of why the verify callback is called multiple
6389 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
6390 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
6391 which causes problems with some clients (such as the Certicom SSL Plus
6392 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
6393 disables the coutermeasure allowing Eudora to connect."
6395 38. Exim was not checking that a write() to a log file succeeded. This could
6396 lead to Bad Things if a log got too big, in particular if it hit a file
6397 size limit. Exim now panics and dies if it cannot write to a log file, just
6398 as it does if it cannot open a log file.
6400 39. Modified OS/Makefile-Linux so that it now contains
6402 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
6404 The two -D definitions ensure that Exim is compiled with large file
6405 support, which makes it possible to handle log files that are bigger than
6408 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
6409 instance) a domain was checked against a named list that involved a lookup,
6410 causing $domain_data to be set, then another domain was checked against the
6411 same list, then the first domain was re-checked, the value of $domain_data
6412 after the final check could be wrong. In particular, if the second check
6413 failed, it could be set empty. This bug probably also applied to
6416 41. The strip_trailing_dot option was not being applied to the address given
6417 with the -f command-line option.
6419 42. The code for reading a message's header from the spool was incrementing
6420 $received_count, but never initializing it. This meant that the value was
6421 incorrect (doubled) while delivering a message in the same process in which
6422 it was received. In the most common configuration of Exim, this never
6423 happens - a fresh exec is done - but it can happen when
6424 deliver_drop_privilege is set.
6426 43. When Exim logs an SMTP synchronization error - client data sent too soon -
6427 it now includes up to 150 characters of the unexpected data in the log
6430 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
6431 and building data strings. The size of both of these buffers was 10 000
6432 bytes - far larger than anybody would *ever* want, thought I. Needless to
6433 say, somebody hit the limit. I have increased the maximum line length to
6434 20 000 and the maximum data length of concatenated lines to 100 000. I have
6435 also fixed two bugs, because there was no checking on these buffers. Tsk,
6436 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
6437 buffer is too small.
6439 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
6440 lsearch lookups. Now it does.
6442 46. When parsing a route_list item in a manualroute router, a fixed-length
6443 buffer was used for the list of hosts. I made this 1024 bytes long,
6444 thinking that nobody would ever have a list of hosts that long. Wrong.
6445 Somebody had a whole pile of complicated expansion conditions, and the
6446 string was silently truncated, leading to an expansion error. It turns out
6447 that it is easier to change to an unlimited length (owing to other changes
6448 that have happened since this code was originally written) than to build
6449 structure for giving a limitation error. The length of the item that
6450 expands into the list of hosts is now unlimited.
6452 47. The lsearch lookup could not handle data where the length of text line was
6453 more than 4095 characters. Such lines were truncated, leading to shortened
6454 data being returned. It should now handle lines of any length.
6456 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
6457 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
6460 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
6461 replacing strings such as BIN_DIRECTORY in a source file: the replacement
6462 no longer happens in comment lines. A list of replacements is now placed
6463 at the head of all of the source files, except those whose only change is
6464 to replace PERL_COMMAND in the very first #! line.
6466 50. Replaced the slow insertion sort in queue.c, for sorting the list of
6467 messages on the queue, with a bottom-up merge sort, using code contributed
6468 by Michael Haardt. This should make operations like -bp somewhat faster on
6469 large queues. It won't affect queue runners, except when queue_run_in_order
6472 51. Installed eximstats 1.31 in the distribution.
6474 52. Added support for SRV lookups to the dnslookup router.
6476 53. If an ACL referred to $message_body or $message_body_end, the value was not
6477 reset for any messages that followed in the same SMTP session.
6479 54. The store-handling optimization for building very long strings was not
6480 differentiating between the different store pools. I don't think this
6481 actually made any difference in practice, but I've tidied it.
6483 55. While running the routers to verify a sender address, $sender_address
6484 was still set to the sender address. This is wrong, because when routing to
6485 send a bounce to the sender, it would be empty. Therefore, I have changed
6486 it so that, while verifying a sender address, $sender_address is set to <>.
6487 (There is no change to what happens when verifying a recipient address.)
6489 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
6490 target A or AAAA records (if not already returned) without resetting the
6491 qualify_single or search_parents options of the DNS resolver. These are
6492 inappropriate in this case because the targets of MX and SRV records must
6493 be FQDNs. A broken DNS record could cause trouble if it happened to have a
6494 target that, when qualified, matched something in the local domain. These
6495 two options are now turned off when doing these lookups.
6497 57. It seems that at least some releases of Reiserfs (which does not have the
6498 concept of a fixed number of inodes) returns zero and not -1 for the
6499 number of available inodes. This interacted badly with check_spool_inodes,
6500 which assumed that -1 was the "no such thing" setting. What I have done is
6501 to check that the total number of inodes is greater than zero before doing
6502 the test of how many are available.
6504 58. When a "warn" ACL statement has a log_message modifier, the message is
6505 remembered, and not repeated. This is to avoid a lot of repetition when a
6506 message has many recipients that cause the same warning to be written.
6507 However, Exim was preserving the list of already written lines for an
6508 entire SMTP session, which doesn't seem right. The memory is now reset if a
6509 new message is started.
6511 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6512 debugging output unless log_rewrite was also set.
6514 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6515 of (int)(handle) when we know that handle contains (void *)(-1).
6517 61. The Exim daemon panic-logs an error return when it closes the incoming
6518 connection. However "connection reset by peer" seems to be common, and
6519 isn't really an error worthy of noting specially, so that particular error
6522 62. When Exim is trying to find all the local interfaces, it used to panic and
6523 die if the ioctl to get the interface flags failed. However, it seems that
6524 on at least one OS (Solaris 9) it is possible to have an interface that is
6525 included in the list of interfaces, but for which you get a failure error
6526 for this call. This happens when the interface is not "plumbed" into a
6527 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6528 failure of the "get flags" call assumes that the interface is down.
6530 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6531 makes life easier for people who are doing arithmetic on fields extracted
6532 from dates, where you often get leading zeros that should not be
6533 interpreted as octal.
6535 64. Added qualify_domain to the redirect router, to override the global
6538 65. If a pathologically long header line contained very many addresses (the
6539 report of this problem mentioned 10 000) and each of them was rewritten,
6540 Exim could use up a very large amount of memory. (It kept on making new
6541 copies of the header line as it rewrote, and never released the old ones.)
6542 At the expense of a bit more processing, the header rewriting function has
6543 been changed so that it no longer eats memory in this way.
6545 66. The generation of the Received: header has been moved from the time that a
6546 message starts to be received, to the time that it finishes. The timestamp
6547 in the Received: header should now be very close to that of the <= log
6548 line. There are two side-effects of this change:
6550 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6551 logged header lines no longer include the local Received: line, because
6552 it has not yet been created. The same applies to a copy of the message
6553 that is returned to a non-SMTP sender when a message is rejected.
6555 (b) When a filter file is tested using -bf, no additional Received: header
6556 is added to the test message. After some thought, I decided that this
6559 This change does not affect the value of $received_for. It is still set
6560 after address rewriting, but before local_scan() is called.
6562 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6564 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6565 gave an unhelpful panic error message, and a defer error. I have managed to
6566 change this behaviour so that it now rejects any supplied certificate,
6567 which seems right, as the list of acceptable certificates is empty.
6569 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6570 gave an unhelpful defer error. I have not managed to make this reject any
6571 supplied certificates, but the error message it gives is "no certificate
6572 supplied", which is not helpful.
6574 70. exigrep's output now also includes lines that are not associated with any
6575 message, but which match the given pattern. Implemented by a patch from
6576 Martin Sluka, which also tidied up the Perl a bit.
6578 71. Recipient callout verification, like sender verification, was using <> in
6579 the MAIL FROM command. This isn't really the right thing, since the actual
6580 sender may affect whether the remote host accepts the recipient or not. I
6581 have changed it to use the actual sender in the callout; this means that
6582 the cache record is now keyed on a recipient/sender pair, not just the
6583 recipient address. There doesn't seem to be a real danger of callout loops,
6584 since a callout by the remote host to check the sender would use <>.
6585 [SEE ABOVE: changed after hitting problems.]
6587 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6588 temporary errors. However, in the case of such a code being given after
6589 the end of a data transmission (i.e. after ".") Exim was failing to write
6590 a retry record for the message. (Yes, there was some broken host that was
6591 actually sending 8xx at this point.)
6593 73. An unknown lookup type in a host list could cause Exim to panic-die when
6594 the list was checked. (An example that provoked this was putting <; in the
6595 middle of a list instead of at the start.) If this happened during a DATA
6596 ACL check, a -D file could be left lying around. This kind of configuration
6597 error no longer causes Exim to die; instead it causes a defer error. The
6598 incident is still logged to the main and panic logs.
6600 74. Buglet left over from Exim 3 conversion. The message "too many messages
6601 in one connection" was written to the rejectlog but not the mainlog, except
6602 when address rewriting (yes!) was being logged.
6604 75. Added write_rejectlog option.
6606 76. When a system filter was run not as root (that is, when system_filter_user
6607 was set), the values of the $n variables were not being returned to the
6608 main process; thus, they were not subsequently available in the $sn
6611 77. Added +return_path_on_delivery log selector.
6613 78. A connection timeout was being treated differently from recipients deferred
6614 when testing hosts_max_try with a message that was older than the host's
6615 retry timeout. (The host should not be counted, thus allowing all hosts to
6616 be tried at least once before bouncing.) This may have been the cause of an
6617 occasionally reported bug whereby a message would remain on the queue
6618 longer than the retry timeout, but would be bounced if a delivery was
6619 forced. I say "may" because I never totally pinned down the problem;
6620 setting up timeout/retry tests is difficult. See also the next item.
6622 79. The ultimate address timeout was not being applied to errors that involved
6623 a combination of host plus message (for example, a timeout on a MAIL
6624 command). When an address resolved to a number of possible hosts, and they
6625 were not all tried for each delivery (e.g. because of hosts_max_try), a
6626 message could remain on the queue longer than the retry timeout.
6628 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6631 81. Fixed an obscure SMTP outgoing bug which required at least the following
6632 conditions: (a) there was another message waiting for the same server;
6633 (b) the server returned 5xx to all RCPT commands in the first message so
6634 that the message was not completed; (c) the server dropped the connection
6635 or gave a negative response to the RSET that Exim sends to abort the
6636 transaction. The observed case was a dropped connection after DATA that had
6637 been sent in pipelining mode. That is, the server had advertised PIPELINING
6638 but was not implementing it correctly. The effect of the bug was incorrect
6639 behaviour, such as trying another host, and this could lead to a crash.
6645 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6646 and daemon.c were passed as pointers to ints; they should have been
6647 pointers to socklen_t variables (which are typically unsigned ints).
6649 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6652 3. Fixed a really odd bug that affected only the testing scheme; patching a
6653 certain fixed string in the binary changed the value of another string that
6654 happened to be identical to the end of the original first string.
6656 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6657 name", it returns that address as the IP address. On some operating
6658 systems (e.g. Solaris), it also passes back the IP address string as the
6659 "host name". However, on others (e.g. Linux), it passes back an empty
6660 string. Exim wasn't checking for this, and was changing the host name to an
6661 empty string, assuming it had been canonicalized.
6663 5. Although rare, it is permitted to have more than one PTR record for a given
6664 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6665 all the names associated with an address, because they do in Solaris.
6666 However, it seems that they do not in Linux for data that comes from the
6667 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6668 I found this out when I moved to a new Linux workstation and tried to run
6669 the Exim test suite.
6671 To get round this problem I have changed the code so that it now does its
6672 own call to the DNS to look up PTR records when searching for a host name.
6673 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6674 addresses that are only in /etc/hosts are still found.
6676 This behaviour is, however, controlled by an option called host_lookup_
6677 order, which defaults to "bydns:byaddr". If people want to use the other
6678 order, or indeed, just use one or the other means of lookup, they can
6679 specify it in this variable.
6681 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6682 some operating systems, this comes back from gethostbyaddr() as an empty
6683 string, and this is what Exim used to test for. However, it seems that in
6684 other systems, "." is yielded. Exim now tests for this case too.
6686 7. The values of check_spool_space and check_log_space are now held internally
6687 as a number of kilobytes instead of an absolute number of bytes. If a
6688 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6689 kilobyte. This means that much larger values can be stored.
6691 8. Exim monitor: an attempt to get the action menu when not actually pointing
6692 at a message produces an empty menu entitled "No message selected". This
6693 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6694 no entries in it ("Shell widget menu has zero width and/or height"). So I
6695 have added a single, blank menu entry in this case.
6697 9. Added ${quote_local_part.
6699 10. MIME decoding is now applied to the contents of Subject: header lines when
6702 11. Now that a reference to $sender_host_address automatically causes a reverse
6703 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6704 host lookup before query-style lookups in lists that might use this
6705 variable. This has therefore been abolished, and the "net-" prefix is no
6706 longer necessary for query-style lookups.
6708 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6709 have been a typo for LFLAGS, so it has been changed.
6711 13. The install script calls Exim with "-C /dev/null" in order to find the
6712 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6713 to be output. However, since Exim outputs its version number before the
6714 error, it didn't break the script. It just looked ugly. I fixed this by
6715 always allowing "-C /dev/null" if the caller is root.
6717 14. Ignore overlarge ACL variable number when reading spool file - insurance
6718 against a later release with more variables having written the file.
6720 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6721 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6722 this, as well as the form without the "IPv6" on the front.
6724 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6725 OS/Makefile-Darwin file.
6727 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6728 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6730 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6731 variable. However, some OS use ints for inodes. I've added cast to long int
6732 to get rid of the compiler warning.
6734 19. I had forgotten to lock out "/../" in configuration file names when
6735 ALT_CONFIG_PREFIX was set.
6737 20. Routers used for verification do not need to specify transports. However,
6738 if such a router generated a host list, and callout was configured, Exim
6739 crashed, because it could not find a port number from the (non-existent)
6740 transport. It now assumes port 25 in this circumstance.
6742 21. Added the -t option to exigrep.
6744 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6745 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6746 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6747 them is set, LOOKUP_LSEARCH is forced.
6749 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6751 24. Added sender and host information to the "rejected by local_scan()" log
6752 line; previously there was no indication of these.
6754 25. Added .include_if_exists.
6756 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6757 Linux. It turns out that not all file systems support this. Apparently some
6758 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6759 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6760 not supported on the file descriptor, is now ignored when Exim is trying to
6761 sync a directory. This applies only to Linux.
6763 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6765 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6766 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6769 29. Added support for maildirsize files from supplied patch (modified a bit).
6771 30. The use of :fail: followed by an empty string could lead Exim to respond to
6772 sender verification failures with (e.g.):
6774 550 Verification failed for <xxx>
6775 550 Sender verify failed
6777 where the first response line was missing the '-' that indicates it is not
6778 the final line of the response.
6780 31. The loop for finding the name of the user that called Exim had a hardwired
6781 limit of 10; it now uses the value of finduser_retries, which is used for
6782 all other user lookups.
6784 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6787 33. Exim was neglecting to zero errno before one call of strtol() when
6788 expanding a string and expecting an integer value. On some systems this
6789 resulted in spurious "integer overflow" errors. Also, it was casting the
6790 result into an int without checking.
6792 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6793 did not work. The code looks as if it has *never* worked, though it appears
6794 to have been documented since at least release 1.62. I have made it work.
6796 35. The "timeout_DNS" error in retry rules, also documented since at least
6797 1.62, also never worked. As it isn't clear exactly what this means, and
6798 clearly it isn't a major issue, I have abolished the feature by treating it
6799 as "timeout", and writing a warning to the main and panic logs.
6801 36. The display of retry rules for -brt wasn't always showing the error code
6804 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6805 timeout_connect_A, timeout_connect_MX.
6807 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6808 to the empty sender.
6810 39. The daemon was not analysing the content of -oX till after it had closed
6811 stderr and disconnected from the controlling terminal. This meant that any
6812 syntax errors were only noted on the panic log, and the return code from
6813 the command was 0. By re-arranging the code a little, I've made the
6814 decoding happen first, so such errors now appear on stderr, and the return
6815 code is 1. However, the actual setting up of the sockets still happens in
6816 the disconnected process, so errors there are still only recorded on the
6819 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6820 connections (as happens on some IP stacks) was logged at start up time as
6821 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6822 it from "IPv6 and IPv4", which means that two separate sockets are being
6825 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6826 says whether AF_INET or AF_INET6 was passed as an argument.
6828 42. Exiwhat output was messed up when time zones were included in log
6831 43. Exiwhat now gives more information about the daemon's listening ports,
6832 and whether -tls-on-connect was used.
6834 44. The "port" option of the smtp transport is now expanded.
6836 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6837 silently ignored. Now an error message is written to the main and panic
6840 46. There's a new ACL modifier called "logwrite" which writes to a log file
6841 as soon as it is encountered.
6843 47. Added $local_user_uid and $local_user_gid at routing time.
6845 48. Exim crashed when trying to verify a sender address that was being
6848 49. Exim was recognizing only a space character after ".include". It now also
6849 recognizes a tab character.
6851 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6852 extracting the relevant information from the specification. The man page no
6853 longer contains scrambled data for the -d option, and I've added a section
6854 at the front about calling Exim under different names.
6856 51. Added "extra_headers" argument to the "mail" command in filter files.
6858 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6861 53. Installed eximstats 1.29.
6863 54. Added transport_filter_timeout as a generic transport option.
6865 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6866 Cc: header lines. This was required by RFC 822, but it not required by RFC
6869 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6870 incoming messages that did not have them. Now it does so only if the
6871 message originates locally, that is, if there is no associated remote host
6872 address. When Resent- header lines are present, this applies to the Resent-
6873 lines rather than the non-Resent- lines.
6875 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6876 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6878 58. Messages for configuration errors now include the name of the main
6879 configuration file - useful now that there may be more than one file in a
6880 list (.included file names were always shown).
6882 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6883 for those rare installations that do not start the daemon as root or run it
6884 setuid root. I've cut out the call to initgroups() if the daemon is not
6887 60. The Exim user and group can now be bound into the binary as text strings
6888 that are looked up at the start of Exim's processing.
6890 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6892 62. Added $mailstore_basename variable.
6894 63. Installed patch to sieve.c from Michael Haardt.
6896 64. When Exim failed to open the panic log after failing to open the main log,
6897 the original message it was trying to log was written to stderr and debug
6898 output, but if they were not available (the usual case in production), it
6899 was lost. Now it is written to syslog before the two lines that record the
6900 failures to open the logs.
6902 65. Users' Exim filters run in subprocesses under the user's uid. It is
6903 possible for a "deliver" command or an alias in a "personal" command to
6904 provoke an address rewrite. If logging of address rewriting is configured,
6905 this fails because the process is not running as root or exim. There may be
6906 a better way of dealing with this, but for the moment (because 4.30 needs
6907 to be released), I have disabled address rewrite logging when running a
6908 filter in a non-root, non-exim process.
6914 1. The buildconfig auxiliary program wasn't quoting the value set for
6915 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6916 not defined. This bug was masked in 4.22 by the effect that was fixed in
6919 2. Some messages that were rejected after a message id was allocated were
6920 shown as "incomplete" by exigrep. It no longer does this for messages that
6921 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6923 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6924 have allow_domain_literals set, the ID did not get logged in the <= line.
6925 Domain literals are now always recognized in Message-ID: header lines.
6927 4. The first argument for a ${extract expansion item is the key name or field
6928 number. Leading and trailing spaces in this item were not being ignored,
6929 causing some misleading effects.
6931 5. When deliver_drop_privilege was set, single queue runner processes started
6932 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6933 same command in the process it spins off) were not dropping privilege.
6935 6. When the daemon running as "exim" started a queue runner, it always
6936 re-executed Exim in the spun-off process. This is a waste of effort when
6937 deliver_drop_privilege is set. The new process now just calls the
6938 queue-runner function directly.
6944 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6947 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6948 ignored. Though the use of -oP was forcing the writing of a pid file, it
6949 was always written to the default place.
6951 3. If the message "no IP address found for host xxxx" is generated during
6952 incoming verification, it is now followed by identification of the incoming
6953 connection (so you can more easily find what provoked it).
6955 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6957 5. Added some features to "harden" Exim a bit more against certain attacks:
6959 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6960 be put in Local/Makefile. This is like the never_users runtime option,
6961 but it cannot be overridden. The default setting is "root".
6963 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6964 prefix string with which any file named in a -C command line option
6967 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6968 is retained for -C and -D only if the caller of Exim is root. Without
6969 it, the exim user may also use -C and -D and retain privilege.
6971 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6972 command line option is disabled.
6974 6. Macro names set by the -D option must start with an upper case letter, just
6975 like macro names defined in the configuration file.
6977 7. Added "dereference=" facility to LDAP.
6979 8. Two instances of the typo "uknown" in the source files are fixed.
6981 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6982 the Configure-Makefile script screwed up while processing it.
6984 10. Incorporated PCRE 4.4.
6986 11. The SMTP synchronization check was not operating right at the start of an
6987 SMTP session. For example, it could not catch a HELO sent before the client
6988 waited for the greeting. There is now a check for outstanding input at the
6989 point when the greeting is written. Because of the duplex, asynchronous
6990 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6991 way, but not yet received, when the check is performed.
6993 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6994 on TCP/IP sockets, because this apparently causes some broken clients to
6997 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6998 unchanged) from the Cygwin maintainer.
7000 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
7001 was supported instead of testing for mbx. Effectively a typo.
7003 15. The spa authenticator server code was not checking that the input it
7004 received was valid base64.
7006 16. The debug output line for the "set" modifier in ACLs was not showing the
7007 name of the variable that was being set.
7009 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
7011 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
7012 lookup on its own. Something else was needed to trigger the lookup. For
7013 example, a match in host_lookup or the need for a host name in a host list.
7014 Now, if $sender_host_name is referenced and the host name has not yet been
7015 looked up, a lookup is performed. If the lookup fails, the variable remains
7016 empty, and $host_lookup_failed is set to "1".
7018 19. Added "eqi" as a case-independent comparison operator.
7020 20. The saslauthd authentication condition could segfault if neither service
7021 nor realm was specified.
7023 21. If an overflowing value such as "2048M" was set for message_size_limit, the
7024 error message that was logged was misleading, and incoming SMTP
7025 connections were dropped. The message is now more accurate, and temporary
7026 errors are given to SMTP connections.
7028 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
7029 (except RSET) with a 421 error, until QUIT is received. However, it was
7030 failing to send a response to QUIT.
7032 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
7033 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
7034 ACL is now run after the helo_try_verify_hosts code.
7036 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
7037 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
7038 case-independent, so other case variants are also recognized). Apparently
7039 some systems use these upper case variants.
7041 25. If more than two messages were waiting for the same host, and a transport
7042 filter was specified for the transport, Exim sent two messages over the
7043 same TCP/IP connection, and then failed with "socket operation on non-
7044 socket" when it tried to send the third.
7046 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
7048 27. The extern definition of crypt16() in expand.c was not being excluded when
7049 the OS had its own crypt16() function.
7051 28. Added bounce_return_body as a new option, and bounce_return_size_limit
7052 as a preferred synonym for return_size_limit, both as an option and as an
7055 29. Added LIBS=-liconv to OS/Makefile-OSF1.
7057 30. Changed the default configuration ACL to relax the local part checking rule
7058 for addresses that are not in any local domains. For these addresses,
7059 slashes and pipe symbols are allowed within local parts, but the sequence
7060 /../ is explicitly forbidden.
7062 31. SPA server authentication was not clearing the challenge buffer before
7065 32. log_message in a "warn" ACL statement was writing to the reject log as
7066 well as to the main log, which contradicts the documentation and doesn't
7067 seem right (because no rejection is happening). So I have stopped it.
7069 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
7070 However, I am unable to do any testing of this.
7072 34. Fixed an infelicity in the appendfile transport. When checking directories
7073 for a mailbox, to see if any needed to be created, it was accidentally
7074 using path names with one or more superfluous leading slashes; tracing
7075 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
7077 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
7078 contents are added to the log line that is written for every discarded
7079 recipient. (Previously a log_message setting was ignored.)
7081 36. The ${quote: operator now quotes the string if it is empty.
7083 37. The install script runs exim in order to find its version number. If for
7084 some reason other than non-existence or emptiness, which it checks, it
7085 could not run './exim', it was installing it with an empty version number,
7086 i.e. as "exim-". This error state is now caught, and the installation is
7089 38. An argument was missing from the function that creates an error message
7090 when Exim fails to connect to the socket for saslauthd authentication.
7091 This could cause Exim to crash, or give a corrupted message.
7093 39. Added isip, isip4, and isip6 to ${if conditions.
7095 40. The ACL variables $acl_xx are now saved with the message, and can be
7096 accessed later in routers, transports, and filters.
7098 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
7099 strings in the file are not string-expanded.
7101 42. If a MAIL command specified a SIZE value that was too large to fit into an
7102 int variable, the check against message_size_limit failed. Such values are
7103 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
7104 day this will have to be increased, but I don't think I want to be around
7105 when emails are that large.
7112 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
7113 iconv() is not standard in FreeBSD.
7115 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
7116 IPv6 enabled. The observed symptom was a segmentation fault on return from
7117 the function os_common_find_running_interfaces() in src/os.c.
7119 3. In the check_special_case() function in daemon.c I had used "errno" as an
7120 argument name, which causes warnings on some systems. This was basically a
7121 typo, since it was named "eno" in the comments!
7123 4. The code that waits for the clock to tick (at a resolution of some fraction
7124 of a second) so as to ensure message-id uniqueness was always waiting for
7125 at least one whole tick, when it could have waited for less. [This is
7126 almost certainly not relevant at current processor speeds, where it is
7127 unlikely to ever wait at all. But we try to future-proof.]
7129 5. The function that sleeps for a time interval that includes fractions of a
7130 second contained a race. It did not block SIGALRM between setting the
7131 timer, and suspending (a couple of lines later). If the interval was short
7132 and the sigsuspend() was delayed until after it had expired, the suspension
7133 never ended. On busy systems this could lead to processes getting stuck for
7136 6. Some uncommon configurations may cause a lookup to happen in a queue runner
7137 process, before it forks any delivery processes. The open lookup caching
7138 mechanism meant that the open file or database connection was passed into
7139 the delivery process. The problem was that delivery processes always tidy
7140 up cached lookup data. This could cause a problem for the next delivery
7141 process started by the queue runner, because the external queue runner
7142 process does not know about the closure. So the next delivery process
7143 still has data in the lookup cache. In the case of a file lookup, there was
7144 no problem because closing a file descriptor in a subprocess doesn't affect
7145 the parent. However, if the lookup was caching a connection to a database,
7146 the connection was closed, and the second delivery process was likely to
7147 see errors such as "PGSQL: query failed: server closed the connection
7148 unexpectedly". The problem has been fixed by closing all cached lookups
7149 in a queue runner before running a delivery process.
7151 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
7152 seem to have the "const" qualifier which it has on other OS. I've
7155 8. Change 4.21/2 was too strict. It is only if there are two authenticators
7156 *of the same type* (client or server) with the same public name that an
7157 error should be diagnosed.
7159 9. When Exim looked up a host name for an IP address, but failed to find the
7160 original IP address when looking up the host name (a safety check), it
7161 output the message "<ip address> does not match any IP for NULL", which was
7162 confusing, to say the least. The bug was that the host name should have
7163 appeared instead of "NULL".
7165 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
7166 user that is built into the binary, and the -C or -D options is used, root
7167 privilege is dropped before the configuration file is read. In addition,
7168 logging is switched to stderr instead of the normal log files. If the
7169 configuration then re-defines the Exim user, the unprivileged environment
7170 is probably not what is expected, so Exim logs a panic warning message (but
7173 However, if deliver_drop_privilege is set, the unprivileged state may well
7174 be exactly what is intended, so the warning has been cut out in that case,
7175 and Exim is allowed to try to write to its normal log files.
7181 1. smtp_return_error_details was not giving details for temporary sender
7182 or receiver verification errors.
7184 2. Diagnose a configuration error if two authenticators have the same public
7187 3. Exim used not to create the message log file for a message until the first
7188 delivery attempt. This could be confusing when incoming messages were held
7189 for policy or load reasons. The message log file is now created at the time
7190 the message is received, and an initial "Received" line is written to it.
7192 4. The automatically generated man page for command line options had a minor
7193 bug that caused no ill effects; however, a more serious problem was that
7194 the procedure for building the man page automatically didn't always
7195 operate. Consequently, release 4.20 contains an out-of-date version. This
7196 shouldn't happen again.
7198 5. When building Exim with embedded Perl support, the script that builds the
7199 Makefile was calling 'perl' to find its compile-time parameters, ignoring
7200 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
7202 6. The freeze_tell option was not being used for messages that were frozen on
7203 arrival, either by an ACL or by local_scan().
7205 7. Added the smtp_incomplete_transaction log selector.
7207 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
7208 was accepting AUTH without a new EHLO.
7210 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
7211 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
7214 10. Logging of TCP/IP connections (when configured) now happens in the main
7215 daemon process instead of the child process, so that the TCP/IP connection
7216 count is more accurate (but it can never be perfect).
7218 11. The use of "drop" in a nested ACL was not being handled correctly in the
7219 outer ACL. Now, if condition failure induced by the nested "drop" causes
7220 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
7221 or "require"), the connection is dropped.
7223 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
7224 that yield "discard" can now be used with an "accept" or a "discard" verb,
7225 but an error is generated for any others (because I can't see a useful way
7226 to define what should happen).
7228 13. When an ACL is read dynamically from a file (or anywhere else), the lines
7229 are now processed in the same way as lines in the Exim configuration file.
7230 In particular, continuation lines are supported.
7232 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
7234 15. Added -ti meaning -t -i.
7236 16. Check for letters, digits, hyphens, and dots in the names of dnslist
7237 domains, and warn by logging if others are found.
7239 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
7240 returned from GIFCONF when Exim is trying to find the list of interfaces on
7241 a host. The code in os.c has been modified to copy each ifreq to an aligned
7242 structure in all cases.
7244 Also, in some cases, the returned ifreq's were being copied to a 'struct
7245 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
7246 means the last couple of bytes of an IPv6 address could be chopped if the
7247 ifreq contained only a normal sockaddr (14 bytes storage).
7249 18. Named domain lists were not supported in the hosts_treat_as_local option.
7250 An entry such as +xxxx was not recognized, and was treated as a literal
7253 19. Ensure that header lines added by a DATA ACL are included in the reject log
7254 if the ACL subsequently rejects the message.
7256 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
7257 MD5 (which is deprecated).
7259 21. When testing a filter file using -bf, Exim was writing a message when it
7260 took the sender from a "From " line in the message, but it was not doing so
7261 when it took $return_path from a Return-Path: header line. It now does.
7263 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
7264 with a valid header line field name (a series of printing characters
7265 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
7267 23. Changed "disc" in the source to "disk" to conform to the documentation and
7268 the book and for uniformity.
7270 24. Ignore Sendmail's -Ooption=value command line item.
7272 25. When execve() failed while trying to run a command in a pipe transport,
7273 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
7274 could be confused with a return value of 69 from the command itself. This
7275 has been changed to 127, the value the shell returns if it is asked to run
7276 a non-existent command. The wording for the related log line suggests a
7277 non-existent command as the problem.
7279 26. If received_header_text expands to an empty string, do not add a Received:
7280 header line to the message. (Well, it adds a token one on the spool, but
7281 marks it "old" so that it doesn't get used or transmitted.)
7283 27. Installed eximstats 1.28 (addition of -nt option).
7285 28. There was no check for failure on the call to getsockname() in the daemon
7286 code. This can fail if there is a shortage of resources on the system, with
7287 ENOMEM, for example. A temporary error is now given on failure.
7289 29. Contrary to the C standard, it seems that in some environments, the
7290 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
7291 program. Exim now does this explicitly; it affects the formatting of
7292 timestamps using strftime().
7294 30. If exiqsumm was given junk data, it threw up some uninitialized variable
7295 complaints. I've now initialized all the variables, to avoid this.
7297 32. Header lines added by a system filter were not being "seen" during
7298 transport-time rewrites.
7300 33. The info_callback() function passed to OpenSSL is set up with type void
7301 (*)(SSL *, int, int), as described somewhere. However, when calling the
7302 function (actually a macro) that sets it up, the type void(*)() is
7303 expected. I've put in a cast to prevent warnings from picky compilers.
7305 34. If a DNS black list lookup found a CNAME record, but there were no A
7306 records associated with the domain it pointed at, Exim crashed.
7308 35. If a DNS black list lookup returned more than one A record, Exim ignored
7309 all but the first. It now scans all returned addresses if a particular IP
7310 value is being sought. In this situation, the contents of the
7311 $dnslist_value variable are a list of all the addresses, separated by a
7314 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
7315 to accept a host name and all its aliases if the forward lookup for any of
7316 them yielded the IP address of the incoming connection. Now it accepts only
7317 those names whose forward lookup yields the correct IP address. Any other
7318 names are discarded. This closes a loophole whereby a rogue DNS
7319 administrator could create reverse DNS records to break through a
7320 wildcarded host restriction in an ACL.
7322 37. If a user filter or a system filter that ran in a subprocess used any of
7323 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
7324 the wrong values were passed to the pipe command ($thisaddress had the
7325 value of $0, $0 had the value of $1, etc). This bug was introduced by
7326 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
7328 38. Improved the line breaking for long SMTP error messages from ACLs.
7329 Previously, if there was no break point between 40 and 75 characters, Exim
7330 left the rest of the message alone. Two changes have been made: (a) I've
7331 reduced the minimum length to 35 characters; (b) if it can't find a break
7332 point between 35 and 75 characters, it looks ahead and uses the first one
7333 that it finds. This may give the occasional overlong line, but at least the
7334 remaining text gets split now.
7336 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
7337 file descriptors might be low, and that setting 1000 would always raise it.
7338 It turns out that in some environments, the limit is already over 1000 and
7339 that lowering it causes trouble. So now Exim takes care not to decrease it.
7341 40. When delivering a message, the value of $return_path is set to $sender_
7342 address at the start of routing (routers may change the value). By an
7343 oversight, this default was not being set up when an address was tested by
7344 -bt or -bv, which affected the outcome if any router or filter referred to
7347 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
7348 log only when "message" or "log_message" are set. However, if one of the
7349 conditions was an address verification, or a call to a nested ACL, the
7350 messages generated by the underlying test were being passed through. This
7351 no longer happens. The underlying message is available in $acl_verify_
7352 message for both "message" and "log_message" expansions, so it can be
7353 passed through if needed.
7355 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
7356 new expansion $bh_ to give the encoded byte string without charset
7357 translation. Translation happens only if iconv() is available; HAVE_ICONV
7358 indicates this at build time. HEADERS_CHARSET gives the charset to
7359 translate to; headers_charset can change it in the configuration, and
7360 "headers charset" can change it in an individual filter file.
7362 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
7363 that creates RFC 2047 encoded "words" labels them as that charset instead
7364 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
7365 expansion operator; (ii) when Exim creates a From: line for a local
7366 message; (iii) when a header line is rewritten to include a "phrase" part.
7368 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
7369 buggy, causing it to skip the first lines of messages whose message ID
7370 ended in 'D'. This would not have bitten before Exim release 4.14, because
7371 message IDs were unlikely to end in 'D' before then. The effect was to have
7372 incorrect size information for certain domains.
7374 45. #include "config.h" was missing at the start of the crypt16.c module. This
7375 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
7378 46. If there was a timeout during a "random" callout check, Exim treated it as
7379 a failure of the random address, and carried on sending RSET and the real
7380 address. If the delay was just some slowness somewhere, the response to the
7381 original RCPT would be taken as a response to RSET and so on, causing
7382 mayhem of various kinds.
7384 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
7385 when I implemented it. It didn't allow for the fact that some option values
7386 may legitimately be negative (e.g. size_addition), and it didn't even do
7387 the right test for positive values.
7389 48. Domain names in DNS records are case-independent. Exim always looks them up
7390 in lower case. Some resolvers return domain names in exactly the case they
7391 appear in the zone file, that is, they may contain uppercase letters. Not
7392 all resolvers do this - some return always lower case. Exim was treating a
7393 change of case by a resolver as a change of domain, similar to a widening
7394 of a domain abbreviation. This triggered its re-routing code and so it was
7395 trying to route what was effectively the same domain again. This normally
7396 caused routing to fail (because the router wouldn't handle the domain
7397 twice). Now Exim checks for this case specially, and just changes the
7398 casing of the domain that it ultimately uses when it transmits the message
7401 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
7404 50. If a filter generated a file delivery with a non-absolute name (possible if
7405 no home directory exists for the router), the forbid_file option was not
7408 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
7409 the existing equality matching.
7411 52. Exim was using ints instead of ino_t variables in some places where it was
7412 dealing with inode numbers.
7414 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
7415 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
7416 called TMPDIR, and if it finds it is different, it changes its value.
7418 54. The smtp_printf() function is now made available to local_scan() so
7419 additional output lines can be written before returning. There is also an
7420 smtp_fflush() function to enable the detection of a dropped connection.
7421 The variables smtp_input and smtp_batched_input are exported to
7424 55. Changed the default runtime configuration: the message "Unknown user"
7425 has been removed from the ACL, and instead placed on the localuser router,
7426 using the cannot_route_message feature. This means that any verification
7427 failures that generate their own messages won't get overridden. Similarly,
7428 the "Unrouteable address" message that was in the ACL for unverifiable
7429 relay addresses has also been removed.
7431 56. Added hosts_avoid_esmtp to the smtp transport.
7433 57. The exicyclog script was not checking for the esoteric option
7434 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
7435 will work only if exicyclog is run under the appropriate euid.
7437 58. Following a discussion on the list, the rules by which Exim recognises line
7438 endings on incoming messages have been changed. The -dropcr and drop_cr
7439 options are now no-ops, retained only for backwards compatibility. The
7440 following line terminators are recognized: LF CRLF CR. However, special
7441 processing applies to CR:
7443 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
7444 nor a local message in the state where . is a terminator.
7446 (ii) If a bare CR is encountered in a header line, an extra space is added
7447 after the line terminator so as not to end the header. The reasoning
7448 behind this is that bare CRs in header lines are most likely either
7449 to be mistakes, or people trying to play silly games.
7451 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
7452 was being incorrectly given as 18 bytes larger than it should have been.
7453 This is a VOB (very old bug).
7455 60. This may never have affected anything current, but just in case it has:
7456 When the local host is found other than at the start of a list of hosts,
7457 the local host, those with the same MX, and any that follow, are discarded.
7458 When the list in question was part of a longer list of hosts, the following
7459 hosts (not currently being processed) were also being discarded. This no
7460 longer happens. I'm not sure if this situation could ever has previously
7463 61. Added the "/MX" feature to lists of hosts in the manualroute and query
7466 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
7467 header. This is something that is recommended in a new Internet Draft, and
7468 is something that is documented as being done by Sendmail. There are two
7469 possible values. For messages generated by the autoreply transport, Exim
7472 Auto-Submitted: auto-replied
7474 whereas for all other generated messages (e.g. bounces) it adds
7476 Auto-Submitted: auto-generated
7478 63. The "personal" condition in filters now includes a test for the
7479 Auto-Submitted: header. If it contains the string "auto-" the message it
7480 not considered personal.
7482 64. Added rcpt_include_affixes as a generic transport option.
7484 65. Added queue_only_override (default true).
7486 66. Added the syslog_duplication option.
7488 67. If what should have been the first header line of a message consisted of
7489 a space followed by a colon, Exim was mis-interpreting it as a header line.
7490 It isn't of course - it is syntactically invalid and should therefore be
7491 treated as the start of the message body. The misbehaviour could have
7492 caused a number of strange effects, including loss of data in subsequent
7493 header lines, and spool format errors.
7495 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
7496 client host had authenticated. This control can now be exercised by an ACL
7497 for more flexibility.
7499 69. By default, callouts do not happen when testing with -bh. There is now a
7500 variant, -bhc, which does actually run the callout code, including
7501 consulting and updating the callout cache.
7503 70. Added support for saslauthd authentication, courtesy of Alexander
7506 71. If statvfs() failed on the spool or log directories while checking their
7507 size for availability, Exim confusingly gave the error "space shortage".
7508 Furthermore, in debugging mode it crashed with a floating point exception.
7509 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7510 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7511 serious problem, Exim now writes to the main and panic logs when this
7512 happens, with details of the failure. It then refuses to accept the
7513 incoming message, giving the message "spool directory problem" or "log
7514 directory problem" with a 421 code for SMTP messages.
7516 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7517 0, 1, and 2 exist, because some OS complain for execs without them (see
7518 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7519 descriptors. However, the code omitted to check that the open succeeded,
7520 causing mysterious errors if for some reason the permissions on /dev/null
7521 got screwed. Now Exim writes a message to the main and panic logs, and
7522 bombs out if it can't open /dev/null.
7524 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7525 interact so that it is all more flexible. It is supposed to remain
7526 backwards compatible. Also added extra_local_interfaces.
7528 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7529 to bomb out with an assertion failure - to the client this appears as a
7530 connection drop. This problem occurs in the part of the code that was taken
7531 from the Samba project. Fortunately, the assertion is in a very simple
7532 function, so I have fixed this by reproducing the function inline in the
7533 one place where it is called, and arranging for authentication to fail
7534 instead of killing the process with assert().
7536 75. The SPA client code was not working when the server requested OEM rather
7537 than Unicode encoding.
7539 76. Added code to make require_files with a specific uid setting more usable in
7540 the case where statting the file as root fails - usually a non-root-mounted
7541 NFS file system. When this happens and the failure is EACCES, Exim now
7542 forks a subprocess and does the per-uid checking as the relevant uid.
7544 77. Added process_log_path.
7546 78. If log_file_path was not explicitly set, a setting of check_log_space or
7547 check_log_inodes was ignored.
7549 79. If a space check for the spool or log partitions fails, the incident is now
7550 logged. Of course, in the latter case the data may get lost...
7552 80. Added the %p formatting code to string_format() so that it can be used to
7553 print addresses in debug_print(). Adjusted all the address printing in the
7554 debugging in store.c to use %p rather than %d.
7556 81. There was a concern that a line of code in smtp_in.c could overflow a
7557 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7558 initially expressed, the concern was not well-founded, because trailing
7559 spaces are removed early. However, if the trailing spaces were followed by
7560 a NULL, they did not get removed, so the overflow was possible. Two fixes
7563 (a) I re-wrote the offending code in a cleaner fashion.
7564 (b) If an incoming SMTP command contains a NULL character, it is rejected
7567 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7568 runs initgroups(), so that if the Exim user is in any additional groups,
7569 they will be used during message reception.
7575 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob