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.
214 JH/01 Remove code calling the customisable local_scan function, unless a new
215 definition "HAVE_LOCAL_SCAN=yes" is present in the Local/Makefile.
217 JH/02 Bug 1007: Avoid doing logging from signal-handlers, as that can result in
218 non-signal-safe functions being used.
220 JH/03 Bug 2269: When presented with a received message having a stupidly large
221 number of DKIM-Signature headers, disable DKIM verification to avoid
222 a resource-consumption attack. The limit is set at twenty.
224 JH/04 Add variables $arc_domains, $arc_oldest_pass for ARC verify. Fix the
225 report of oldest_pass in ${authres } in consequence, and separate out
226 some descriptions of reasons for verification fail.
228 JH/05 Bug 2273: Cutthrough delivery left a window where the received messsage
229 files in the spool were present and unlocked. A queue-runner could spot
230 them, resulting in a duplicate delivery. Fix that by doing the unlock
231 after the unlink. Investigation by Tim Stewart. Take the opportunity to
232 add more error-checking on spoolfile handling while that code is being
235 PP/01 Refuse to open a spool data file (*-D) if it's a symlink.
236 No known attacks, no CVE, this is defensive hardening.
238 JH/06 Bug 2275: The MIME ACL unlocked the received message files early, and
239 a queue-runner could start a delivery while other operations were ongoing.
240 Cutthrough delivery was a common victim, resulting in duplicate delivery.
241 Found and investigated by Tim Stewart. Fix by using the open message data
242 file handle rather than opening another, and not locally closing it (which
243 releases a lock) for that case, while creating the temporary .eml format
244 file for the MIME ACL. Also applies to "regex" and "spam" ACL conditions.
246 JH/07 Bug 177: Make a random-recipient callout success visible in ACL, by setting
247 $sender_verify_failure/$recipient_verify_failure to "random".
249 JH/08 When generating a selfsigned cert, use serial number 1 since zero is not
252 JH/09 Bug 2274: Fix logging of cmdline args when starting in an unlinked cwd.
253 Previously this would segfault.
255 JH/10 Fix ARC signing for case when DKIM signing failed. Previously this would
258 JH/11 Bug 2264: Exim now only follows CNAME chains one step by default. We'd
259 like zero, since the resolver should be doing this for us, But we need one
260 as a CNAME but no MX presence gets the CNAME returned; we need to check
261 that doesn't point to an MX to declare it "no MX returned" rather than
262 "error, loop". A new main option is added so the older capability of
263 following some limited number of chain links is maintained.
265 JH/12 Add client-ip info to non-pass iprev ${authres } lines.
267 JH/13 For receent Openssl versions (1.1 onward) use modern generic protocol
268 methods. These should support TLS 1.3; they arrived with TLS 1.3 and the
269 now-deprecated earlier definitions used only specified the range up to TLS
270 1.2 (in the older-version library docs).
272 JH/14 Bug 2284: Fix DKIM signing for body lines starting with a pair of dots.
274 JH/15 Rework TLS client-side context management. Stop using a global, and
275 explicitly pass a context around. This enables future use of TLS for
276 connections to service-daemons (eg. malware scanning) while a client smtp
277 connection is using TLS; with cutthrough connections this is quite likely.
279 JH/16 Fix ARC verification to do AS checks in reverse order.
281 JH/17 Support a "tls" option on the ${readsocket } expansion item.
283 JH/18 Bug 2287: Fix the protocol name (eg utf8esmtp) for multiple messages
284 using the SMTPUTF8 option on their MAIL FROM commands, in one connection.
285 Previously the "utf8" would be re-prepended for every additional message.
287 JH/19 Reject MAIL FROM commands with SMTPUTF8 when the facility was not advertised.
288 Previously thery were accepted, resulting in issues when attempting to
289 forward messages to a non-supporting MTA.
291 PP/02 Let -n work with printing macros too, not just options.
293 JH/20 Bug 2296: Fix cutthrough for >1 address redirection. Previously only
294 one parent address was copied, and bogus data was used at delivery-logging
295 time. Either a crash (after delivery) or bogus log data could result.
296 Discovery and analysis by Tim Stewart.
298 PP/03 Make ${utf8clean:} expansion operator detect incomplete final character.
299 Previously if the string ended mid-character, we did not insert the
300 promised '?' replacement.
302 PP/04 Documentation: current string operators work on bytes, not codepoints.
304 JH/21 Change as many as possible of the global flags into one-bit bitfields; these
305 should pack well giving a smaller memory footprint so better caching and
306 therefore performance. Group the declarations where this can't be done so
307 that the byte-sized flag variables are not interspersed among pointer
308 variables, giving a better chance of good packing by the compiler.
310 JH/22 Bug 1896: Fix the envelope from for DMARC forensic reports to be possibly
311 non-null, to avoid issues with sites running BATV. Previously reports were
312 sent with an empty envelope sender so looked like bounces.
314 JH/23 Bug 2318: Fix the noerror command within filters. It wasn't working.
315 The ignore_error flag wasn't being returned from the filter subprocess so
316 was not set for later routers. Investigation and fix by Matthias Kurz.
318 JH/24 Bug 2310: Raise a msg:fail:internal event for each undelivered recipient,
319 and a msg:complete for the whole, when a message is manually removed using
320 -Mrm. Developement by Matthias Kurz, hacked on by JH.
322 JH/25 Avoid fixed-size buffers for pathnames in DB access. This required using
323 a "Gnu special" function, asprintf() in the DB utility binary builds; I
324 hope that is portable enough.
326 JH/26 Bug 2311: Fix DANE-TA verification under GnuTLS. Previously it was also
327 requiring a known-CA anchor certificate; make it now rely entirely on the
328 TLSA as an anchor. Checking the name on the leaf cert against the name
329 on the A-record for the host is still done for TA (but not for EE mode).
331 JH/27 Fix logging of proxy address. Previously, a pointless "PRX=[]:0" would be
332 included in delivery lines for non-proxied connections, when compiled with
333 SUPPORT_SOCKS and running with proxy logging enabled.
335 JH/28 Bug 2314: Fire msg:fail:delivery event even when error is being ignored.
336 Developement by Matthias Kurz, tweaked by JH. While in that bit of code,
337 move the existing event to fire before the normal logging of message
338 failure so that custom logging is bracketed by normal logging.
340 JH/29 Bug 2322: A "fail" command in a non-system filter (file) now fires the
341 msg:fail:internal event. Developement by Matthias Kurz.
343 JH/30 Bug 2329: Increase buffer size used for dns lookup from 2k, which was
344 far too small for todays use of crypto signatures stored there. Go all
345 the way to the max DNS message size of 64kB, even though this might be
346 overmuch for IOT constrained device use.
348 JH/31 Fix a bad use of a copy function, which could be used to pointlessly
349 copy a string over itself. The library routine is documented as not
350 supporting overlapping copies, and on MacOS it actually raised a SIGABRT.
352 JH/32 For main options check_spool_space and check_inode_space, where the
353 platform supports 64b integers, support more than the previous 2^31 kB
354 (i.e. more than 2 TB). Accept E, P and T multipliers in addition to
355 the previous G, M, k.
357 JH/33 Bug 2338: Fix the cyrus-sasl authenticator to fill in the
358 $authenticated_fail_id variable on authentication failure. Previously
361 JH/34 Increase RSA keysize of autogen selfsign cert from 1024 to 2048. RHEL 8.0
362 OpenSSL didn't want to use such a weak key. Do for GnuTLS also, and for
363 more-modern GnuTLS move from GNUTLS_SEC_PARAM_LOW to
364 GNUTLS_SEC_PARAM_MEDIUM.
366 JH/35 OpenSSL: fail the handshake when SNI processing hits a problem, server
367 side. Previously we would continue as if no SNI had been received.
369 JH/36 Harden the handling of string-lists. When a list consisted of a sole
370 "<" character, which should be a list-separator specification, we walked
371 off past the nul-terimation.
373 JH/37 Bug 2341: Send "message delayed" warning MDNs (restricted to external
374 causes) even when the retry time is not yet met. Previously they were
375 not, meaning that when (say) an account was over-quota and temp-rejecting,
376 and multiple senders' messages were queued, only one sender would get
377 notified on each configured delay_warning cycle.
379 JH/38 Bug 2351: Log failures to extract envelope addresses from message headers.
381 JH/39 OpenSSL: clear the error stack after an SSL_accept(). With anon-auth
382 cipher-suites, an error can be left on the stack even for a succeeding
383 accept; this results in impossible error messages when a later operation
386 AM/01 Bug 2359: GnuTLS: repeat lowlevel read and write operations while they
387 return error codes indicating retry. Under TLS1.3 this becomes required.
389 JH/40 Fix the feature-cache refresh for EXPERIMENTAL_PIPE_CONNECT. Previously
390 it only wrote the new authenticators, resulting in a lack of tracking of
391 peer changes of ESMTP extensions until the next cache flush.
393 JH/41 Fix the loop reading a message header line to check for integer overflow,
394 and more-often against header_maxsize. Previously a crafted message could
395 induce a crash of the recive process; now the message is cleanly rejected.
397 JH/42 Bug 2366: Fix the behaviour of the dkim_verify_signers option. It had
398 been totally disabled for all of 4.91. Discovery and fix by "Mad Alex".
404 GF/01 DEFER rather than ERROR on redis cluster MOVED response.
405 When redis_servers is set to a list of > 1 element, and the Redis servers
406 in that list are in cluster configuration, convert the REDIS_REPLY_ERROR
407 case of MOVED into a DEFER case instead, thus moving the query onto the
408 next server in the list. For a cluster of N elements, all N servers must
409 be defined in redis_servers.
411 GF/02 Catch and remove uninitialized value warning in exiqsumm
412 Check for existence of @ARGV before looking at $ARGV[0]
414 JH/01 Replace the store_release() internal interface with store_newblock(),
415 which internalises the check required to safely use the old one, plus
416 the allocate and data copy operations duplicated in both (!) of the
417 extant use locations.
419 JH/02 Disallow '/' characters in queue names specified for the "queue=" ACL
420 modifier. This matches the restriction on the commandline.
422 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
423 Previously only the last row was returned.
425 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
426 we assumed that tags in the header were well-formed, and parsed the
427 element content after inspecting only the first char of the tag.
428 Assumptions at that stage could crash the receive process on malformed
431 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
432 While running the DKIM ACL we operate on the Permanent memory pool so that
433 variables created with "set" persist to the DATA ACL. Also (at any time)
434 DNS lookups that fail create cache records using the Permanent pool. But
435 expansions release any allocations made on the current pool - so a dnsdb
436 lookup expansion done in the DKIM ACL releases the memory used for the
437 DNS negative-cache, and bad things result. Solution is to switch to the
438 Main pool for expansions.
439 While we're in that code, add checks on the DNS cache during store_reset,
440 active in the testsuite.
441 Problem spotted, and debugging aided, by Wolfgang Breyha.
443 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
444 When none of the hosts presented to a transport match an already-open
445 connection, close it and proceed with the list. Previously we would
446 queue the message. Spotted by Lena with Yahoo, probably involving
449 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
450 Previously a spurious "250 OK id=" response was appended to the proper
453 JH/08 The "support for" informational output now, which built with Content
454 Scanning support, has a line for the malware scanner interfaces compiled
455 in. Interface can be individually included or not at build time.
457 JH/09 The "aveserver", "kavdaemon" and "mksd" interfaces are now not included
458 by the template makefile "src/EDITME". The "STREAM" support for an older
459 ClamAV interface method is removed.
461 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
462 rows affected is given instead).
464 JH/11 The runtime Berkeley DB library version is now additionally output by
465 "exim -d -bV". Previously only the compile-time version was shown.
467 JH/12 Bug 2230: Fix cutthrough routing for nonfirst messages in an initiating
468 SMTP connection. Previously, when one had more receipients than the
469 first, an abortive onward connection was made. Move to full support for
470 multiple onward connections in sequence, handling cutthrough connection
471 for all multi-message initiating connections.
473 JH/13 Bug 2229: Fix cutthrough routing for nonstandard port numbers defined by
474 routers. Previously, a multi-recipient message would fail to match the
475 onward-connection opened for the first recipient, and cause its closure.
477 JH/14 Bug 2174: A timeout on connect for a callout was also erroneously seen as
478 a timeout on read on a GnuTLS initiating connection, resulting in the
479 initiating connection being dropped. This mattered most when the callout
480 was marked defer_ok. Fix to keep the two timeout-detection methods
483 JH/15 Relax results from ACL control request to enable cutthrough, in
484 unsupported situations, from error to silently (except under debug)
485 ignoring. This covers use with PRDR, frozen messages, queue-only and
488 HS/01 Fix Buffer overflow in base64d() (CVE-2018-6789)
490 JH/16 Fix bug in DKIM verify: a buffer overflow could corrupt the malloc
491 metadata, resulting in a crash in free().
493 PP/01 Fix broken Heimdal GSSAPI authenticator integration.
494 Broken in f2ed27cf5, missing an equals sign for specified-initialisers.
495 Broken also in d185889f4, with init system revamp.
497 JH/17 Bug 2113: Fix conversation closedown with the Avast malware scanner.
498 Previously we abruptly closed the connection after reading a malware-
499 found indication; now we go on to read the "scan ok" response line,
502 JH/18 Bug 2239: Enforce non-usability of control=utf8_downconvert in the mail
503 ACL. Previously, a crash would result.
505 JH/19 Speed up macro lookups during configuration file read, by skipping non-
506 macro text after a replacement (previously it was only once per line) and
507 by skipping builtin macros when searching for an uppercase lead character.
509 JH/20 DANE support moved from Experimental to mainline. The Makefile control
510 for the build is renamed.
512 JH/21 Fix memory leak during multi-message connections using STARTTLS. A buffer
513 was allocated for every new TLS startup, meaning one per message. Fix
514 by only allocating once (OpenSSL) or freeing on TLS-close (GnuTLS).
516 JH/22 Bug 2236: When a DKIM verification result is overridden by ACL, DMARC
517 reported the original. Fix to report (as far as possible) the ACL
518 result replacing the original.
520 JH/23 Fix memory leak during multi-message connections using STARTTLS under
521 OpenSSL. Certificate information is loaded for every new TLS startup,
522 and the resources needed to be freed.
524 JH/24 Bug 2242: Fix exim_dbmbuild to permit directoryless filenames.
526 JH/25 Fix utf8_downconvert propagation through a redirect router. Previously it
529 JH/26 Bug 2253: For logging delivery lines under PRDR, append the overall
530 DATA response info to the (existing) per-recipient response info for
531 the "C=" log element. It can have useful tracking info from the
532 destination system. Patch from Simon Arlott.
534 JH/27 Bug 2251: Fix ldap lookups that return a single attribute having zero-
535 length value. Previously this would segfault.
537 HS/02 Support Avast multiline protoocol, this allows passing flags to
538 newer versions of the scanner.
540 JH/28 Ensure that variables possibly set during message acceptance are marked
541 dead before release of memory in the daemon loop. This stops complaints
542 about them when the debug_store option is enabled. Discovered specifically
543 for sender_rate_period, but applies to a whole set of variables.
544 Do the same for the queue-runner and queue-list loops, for variables set
545 from spool message files. Do the same for the SMTP per-message loop, for
546 certain variables indirectly set in ACL operations.
548 JH/29 Bug 2250: Fix a longstanding bug in heavily-pipelined SMTP input (such
549 as a multi-recipient message from a mailinglist manager). The coding had
550 an arbitrary cutoff number of characters while checking for more input;
551 enforced by writing a NUL into the buffer. This corrupted long / fast
552 input. The problem was exposed more widely when more pipelineing of SMTP
553 responses was introduced, and one Exim system was feeding another.
554 The symptom is log complaints of SMTP syntax error (NUL chars) on the
555 receiving system, and refused recipients seen by the sending system
556 (propating to people being dropped from mailing lists).
557 Discovered and pinpointed by David Carter.
559 JH/30 The (EXPERIMENTAL_DMARC) variable $dmarc_ar_header is withdrawn, being
560 replaced by the ${authresults } expansion.
562 JH/31 Bug 2257: Fix pipe transport to not use a socket-only syscall.
564 HS/03 Set a handler for SIGTERM and call exit(3) if running as PID 1. This
565 allows proper process termination in container environments.
567 JH/32 Bug 2258: Fix spool_wireformat in combination with LMTP transport.
568 Previously the "final dot" had a newline after it; ensure it is CR,LF.
570 JH/33 SPF: remove support for the "spf" ACL condition outcome values "err_temp"
571 and "err_perm", deprecated since 4.83 when the RFC-defined words
572 "temperror" and "permerror" were introduced.
574 JH/34 Re-introduce enforcement of no cutthrough delivery on transports having
575 transport-filters or DKIM-signing. The restriction was lost in the
576 consolidation of verify-callout and delivery SMTP handling.
577 Extend the restriction to also cover ARC-signing.
579 JH/35 Cutthrough: for a final-dot response timeout (and nonunderstood responses)
580 in defer=pass mode supply a 450 to the initiator. Previously the message
583 PP/02 DANE: add dane_require_tls_ciphers SMTP Transport option; if unset,
584 tls_require_ciphers is used as before.
586 HS/03 Malware Avast: Better match the Avast multiline protocol. Add
587 "pass_unscanned". Only tmpfails from the scanner are written to
588 the paniclog, as they may require admin intervention (permission
589 denied, license issues). Other scanner errors (like decompression
590 bombs) do not cause a paniclog entry.
592 JH/36 Fix reinitialisation of DKIM logging variable between messages.
593 Previously it was possible to log spurious information in receive log
596 JH/37 Bug 2255: Revert the disable of the OpenSSL session caching. This
597 triggered odd behaviour from Outlook Express clients.
599 PP/03 Add util/renew-opendmarc-tlds.sh script for safe renewal of public
602 JH/38 DKIM: accept Ed25519 pubkeys in SubjectPublicKeyInfo-wrapped form,
603 since the IETF WG has not yet settled on that versus the original
604 "bare" representation.
606 JH/39 Fix syslog logging for syslog_timestamp=no and log_selector +millisec.
607 Previously the millisecond value corrupted the output.
608 Fix also for syslog_pid=no and log_selector +pid, for which the pid
609 corrupted the output.
615 JH/01 Rework error string handling in TLS interface so that the caller in
616 more cases is responsible for logging. This permits library-sourced
617 string to be attached to addresses during delivery, and collapses
618 pairs of long lines into single ones.
620 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
621 during configuration. Wildcards are allowed and expanded.
623 JH/02 Rework error string handling in DKIM to pass more info back to callers.
624 This permits better logging.
626 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
627 do not close it down and have the child transport start it up again on
628 the passed-on TCP connection. Instead, proxy the child (and any
629 subsequent ones) for TLS via a unix-domain socket channel. Logging is
630 affected: the continued delivery log lines do not have any DNSSEC, TLS
631 Certificate or OCSP information. TLS cipher information is still logged.
633 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
634 identical IP addresses on different listening ports. Will also affect
637 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
638 add noisy ifdef guards to special-case this sillyness.
639 Patch from Bernd Kuhls.
641 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
642 than 255 are no longer allowed.
644 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
645 Disable the session-cache too, which might reduce our load. Since we
646 currrectly use a new context for every connection, both as server and
647 client, there is no benefit for these.
648 GnuTLS appears to not support tickets server-side by default (we don't
649 call gnutls_session_ticket_enable_server()) but client side is enabled
650 by default on recent versions (3.1.3 +) unless the PFS priority string
653 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
654 <https://reproducible-builds.org/specs/source-date-epoch/>.
656 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
657 the check for any unsuccessful recipients did not notice the limit, and
658 erroneously found still-pending ones.
660 JH/08 Pipeline CHUNKING command and data together, on kernels that support
661 MSG_MORE. Only in-clear (not on TLS connections).
663 JH/09 Avoid using a temporary file during transport using dkim. Unless a
664 transport-filter is involved we can buffer the headers in memory for
665 creating the signature, and read the spool data file once for the
666 signature and again for transmission.
668 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
669 4.77 as the kernel support then wasn't solid, having issues in 64bit
670 mode. Now, it's been long enough. Add support for FreeBSD also.
672 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
673 case where the routing stage had gathered several addresses to send to
674 a host before calling the transport for the first, we previously failed
675 to close down TLS in the old transport process before passing the TCP
676 connection to the new process. The new one sent a STARTTLS command
677 which naturally failed, giving a failed delivery and bloating the retry
678 database. Investigation and fix prototype from Wolfgang Breyha.
680 JH/12 Fix check on SMTP command input synchronisation. Previously there were
681 false-negatives in the check that the sender had not preempted a response
682 or prompt from Exim (running as a server), due to that code's lack of
683 awareness of the SMTP input buffering.
685 PP/04 Add commandline_checks_require_admin option.
686 Exim drops privileges sanely, various checks such as -be aren't a
687 security problem, as long as you trust local users with access to their
688 own account. When invoked by services which pass untrusted data to
689 Exim, this might be an issue. Set this option in main configuration
690 AND make fixes to the calling application, such as using `--` to stop
693 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
694 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
695 responses to those, into a single TLS record each way (this usually means
696 a single packet). As a side issue, smtp_enforce_sync now works on TLS
699 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
700 affects you only if you're dancing at the edge of the param size limits.
701 If you are, and this message makes sense to you, then: raise the
702 configured limit or use OpenSSL 1.1. Nothing we can do for older
705 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
706 cmdline element to get the documented default one. Previously it was
709 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
710 are unsuable for non-message-specific reasons - eg. connection timeout,
711 banner-time rejection.
713 JH/16 Fix logging of delivery remote port, when specified by router, under
716 PP/06 Repair manualroute's ability to take options in any order, even if one
717 is the name of a transport.
720 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
722 JH/17 Change the list-building routines interface to use the expanding-string
723 triplet model, for better allocation and copying behaviour.
725 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
726 Previously it was constructed the first time a possibly-matching string
727 was met in the configuration file input during startup; now it is done
730 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
731 compatible one, to avoid the (poorly documented) possibility of a config
732 file in the working directory redirecting the DB files, possibly correpting
733 some existing file. CVE-2017-10140 assigned for BDB.
735 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
736 cache-hot. Previously, although the result was properly cached, the
737 initial verify call returned a defer.
739 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
740 the main verify for receipient in uncached-mode.
742 JH/22 Retire historical build files to an "unsupported" subdir. These are
743 defined as "ones for which we have no current evidence of testing".
745 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
746 if present. Previously it was ignored.
748 JH/24 Start using specified-initialisers in C structure init coding. This is
749 a C99 feature (it's 2017, so now considered safe).
751 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
752 if was a fixed-sized field and bitmask ops via macros; it is now more
755 PP/07 GitHub PR 56: Apply MariaDB build fix.
756 Patch provided by Jaroslav Škarvada.
758 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
759 during Coverity cleanups [4.87 JH/47]
760 Diagnosis and fix provided by Michael Fischer v. Mollard.
762 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
763 the right size to place the terminating semicolon on its own folded
764 line, the header hash was calculated to an incorrect value thanks to
765 the (relaxed) space the fold became.
767 HS/02 Fix Bug 2130: large writes from the transport subprocess were chunked
768 and confused the parent.
770 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
771 which could crash as a result. This could lead to undeliverable messages.
773 JH/28 Logging: "next input sent too soon" now shows where input was truncated
776 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
777 matters on fast-turnover and PID-randomising systems, which were getting
778 out-of-order delivery.
780 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
781 a possibly-overlapping copy. The symptom was that "Remote host closed
782 connection in response to HELO" was logged instead of the actual 4xx
785 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
786 Previously only that bufferd was discarded, resulting in SYMTP command
789 JH/32 DKIM: when a message has multiple signatures matching an identity given
790 in dkim_verify_signers, run the dkim acl once for each. Previously only
791 one run was done. Bug 2189.
793 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
794 "panic the current process" to "deliberately defer". The panic log is
795 still written with the problem list name; the mail and reject logs now
796 get a temp-reject line for the message that was being handled, saying
797 something like "domains check lookup or other defer". The SMTP 451
798 message is still "Temporary local problem".
800 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
801 A crafted sequence of BDAT commands could result in in-use memory beeing
802 freed. CVE-2017-16943.
804 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
805 from SMTP input. Previously it was always done; now only done for DATA
806 and not BDAT commands. CVE-2017-16944.
808 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
809 to the message (such as an overlong header line). Previously this was
810 not done and we did not exit BDAT mode. Followon from the previous item
811 though a different problem.
817 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
818 than -2003 did; needs libidn2 in addition to libidn.
820 JH/02 The path option on a pipe transport is now expanded before use.
822 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
823 Patch provided by "Björn", documentation fix added too.
825 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
826 missing a wire-to-host endian conversion.
828 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
829 close after a BDAT command line could be taken as a following command,
830 giving a synch failure. Fix by only checking for synch immediately
831 before acknowledging the chunk.
833 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
834 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
835 macro. Patches provided by Josh Soref.
837 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
838 Previously we did not; the RFC seems ambiguous and VRFY is not listed
839 by IANA as a service extension. However, John Klensin suggests that we
842 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
843 the dkim code may be unix-mode line endings rather than smtp wire-format
844 CRLF, so prepend a CR to any bare LF.
846 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
847 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
849 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
850 the first were themselves being wrongly included in the feed into dkim
851 processing; with most chunk sizes in use this resulted in an incorrect
852 body hash calculated value.
854 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
855 DKIM signature block, for verification. Although advised against by
856 standards it is specifically not ruled illegal.
858 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
860 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
861 missing a body hash (the bh= tag).
863 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
864 It seems that HAProxy sends the Proxy Protocol information in clear and
865 only then does a TLS startup, so do the same.
867 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
868 TCP connections (such as for Spamd) unless the daemon successfully set
869 Fast Open mode on its listening sockets. This fixes breakage seen on
870 too-old kernels or those not configured for Fast Open, at the cost of
871 requiring both directions being enabled for TFO, and TFO never being used
872 by non-daemon-related Exim processes.
874 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
875 endings, at least on the first header line. Try to canonify any that get
876 past that check, despite the cost.
878 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
879 now limited to an arbitrary five deep, while parsing addresses with the
880 strip_excess_angle_brackets option enabled.
882 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
883 instead leave the unprompted TLS handshake in socket buffer for the
884 TLS library to consume.
886 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
888 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
890 JH/16 Drop variables when they go out of scope. Memory management drops a whole
891 region in one operation, for speed, and this leaves assigned pointers
892 dangling. Add checks run only under the testsuite which checks all
893 variables at a store-reset and panics on a dangling pointer; add code
894 explicitly nulling out all the variables discovered. Fixes one known
895 bug: a transport crash, where a dangling pointer for $sending_ip_address
896 originally assigned in a verify callout, is re-used.
898 PP/06 Drop '.' from @INC in various Perl scripts.
900 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
902 PP/08 Reduce a number of compilation warnings under clang; building with
903 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
904 should be warning-free.
906 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
908 HS/01 Fix portability problems introduced by PP/08 for platforms where
909 realloc(NULL) is not equivalent to malloc() [SunOS et al].
911 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
912 chunk. This allows us to accept broken chunked messages. We need a more
913 general solution here.
915 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
916 already-broken messages in the queue.
918 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
920 JH/19 Fix reference counting bug in routing-generated-address tracking.
926 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
927 supports it and a size is available (ie. the sending peer gave us one).
929 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
930 years of being deprecated). The replacements are the ACLs
931 acl_smtp_mime and acl_not_smtp_mime.
933 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
934 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
935 or even an in-clear connection were permitted. Now, if the host lookup
936 was dnssec and dane was requested then the host is only used if the
937 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
938 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
939 if one fails this test.
940 This means that a poorly-configured remote DNS will make it incommunicado;
941 but it protects against a DNS-interception attack on it.
943 JH/04 Bug 1810: make continued-use of an open smtp transport connection
944 non-noisy when a race steals the message being considered.
946 JH/05 If main configuration option tls_certificate is unset, generate a
947 self-signed certificate for inbound TLS connections.
949 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
950 in rewrites and routers.
952 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
953 and logged a warning sing 4.83; now they are a configuration file error.
955 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
956 (lacking @domain). Apply the same qualification processing as RCPT.
958 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
960 JH/10 Support ${sha256:} applied to a string (as well as the previous
963 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
964 a cutthrough deliver is pending, as we always want to make a connection.
965 This also avoids re-routing the message when later placing the cutthrough
966 connection after a verify cache hit.
967 Do not update it with the verify result either.
969 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
970 when routing results in more than one destination address.
972 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
973 signing (which inhibits the cutthrough capability). Previously only
974 the presence of an option was tested; now an expansion evaluating as
975 empty is permissible (obviously it should depend only on data available
976 when the cutthrough connection is made).
978 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
979 the relevant preceding SMTP command did not note the pipelining mode.
981 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
982 Previously they were not counted.
984 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
985 as one having no matching records. Previously we deferred the message
986 that needed the lookup.
988 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
989 distinguished as "(=".
991 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
992 for missing MX records. Previously it only worked for missing A records.
994 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
996 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
997 after the data-go-ahead and data-ack. Patch from Jason Betts.
999 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
1000 even for a "none" policy. Patch from Tony Meyer.
1002 JH/22 Fix continued use of a connection for further deliveries. If a port was
1003 specified by a router, it must also match for the delivery to be
1006 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
1007 When one of the recipients of a message was unsuitable for the connection
1008 (has no matching addresses), we lost track of needing to mark it
1009 deferred. As a result mail would be lost.
1011 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
1013 JH/25 Decoding ACL controls is now done using a binary search; the source code
1014 takes up less space and should be simpler to maintain. Merge the ACL
1015 condition decode tables also, with similar effect.
1017 JH/26 Fix problem with one_time used on a redirect router which returned the
1018 parent address unchanged. A retry would see the parent address marked as
1019 delivered, so not attempt the (identical) child. As a result mail would
1022 JH/27 Fix a possible security hole, wherein a process operating with the Exim
1023 UID can gain a root shell. Credit to http://www.halfdog.net/ for
1024 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
1027 JH/28 Enable {spool,log} filesystem space and inode checks as default.
1028 Main config options check_{log,spool}_{inodes,space} are now
1029 100 inodes, 10MB unless set otherwise in the configuration.
1031 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
1032 Previously it only applied to the main-section connection policy
1035 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
1037 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
1038 by me. Added RFC7919 DH primes as an alternative.
1040 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
1041 are not in the system include path.
1043 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
1044 GnuTLS, when a session startup failed (eg because the client disconnected)
1045 Exim did stdio operations after fclose. This was exposed by a recent
1046 change which nulled out the file handle after the fclose.
1048 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
1049 signed directly by the cert-signing cert, rather than an intermediate
1050 OCSP-signing cert. This is the model used by LetsEncrypt.
1052 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
1054 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
1055 an incoming connection.
1057 HS/02 Bug 1802: Do not half-close the connection after sending a request
1060 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
1061 fallback to "prime256v1".
1063 JH/34 SECURITY: Use proper copy of DATA command in error message.
1064 Could leak key material. Remotely exploitable. CVE-2016-9963.
1070 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
1071 and 3.4.4 - once the server is enabled to respond to an OCSP request
1072 it does even when not requested, resulting in a stapling non-aware
1073 client dropping the TLS connection.
1075 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
1076 support variable-length bit vectors. No functional change.
1078 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
1079 The I= interface field on outgoing lines is now after the H= remote
1080 host field, same as incoming lines. There is a separate
1081 outgoing_interface log selector which allows you to disable the
1084 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
1085 If not running log_selector +smtp_connection the mainlog would be held
1086 open indefinitely after a "too many connections" event, including to a
1087 deleted file after a log rotate. Leave the per net connection logging
1088 leaving it open for efficiency as that will be quickly detected by the
1089 check on the next write.
1091 HS/01 Bug 1671: Fix post transport crash.
1092 Processing the wait-<transport> messages could crash the delivery
1093 process if the message IDs didn't exist for some reason. When
1094 using 'split_spool_directory=yes' the construction of the spool
1095 file name failed already, exposing the same netto behaviour.
1097 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
1098 mime_regex ACL conditions.
1100 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
1101 to DSN fail messages (bounces): remote IP, remote greeting, remote response
1102 to HELO, local diagnostic string.
1104 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
1105 log line to debug. Even when configured with a tls authenticator many
1106 client connections are expected to not authenticate in this way, so
1107 an authenticate fail is not an error.
1109 HS/02 Add the Exim version string to the process info. This way exiwhat
1110 gives some more detail about the running daemon.
1112 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
1113 matter for fast-change records such as DNSBLs.
1115 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
1116 retry record, even if constant. There may be multiple transports with
1117 different interface settings and the retry behaviour needs to be kept
1120 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
1122 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
1124 JH/10 Bug 840: fix log_defer_output option of pipe transport
1126 JH/11 Bug 830: use same host for all RCPTS of a message, even under
1127 hosts_randomize. This matters a lot when combined with mua_wrapper.
1129 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
1130 ${quote_pgsql:<string>} operator.
1132 JH/13 Bug 1708: avoid misaligned access in cached lookup.
1134 JH/14 Change header file name for freeradius-client. Relevant if compiling
1135 with Radius support; from the Gentoo tree and checked under Fedora.
1137 JH/15 Bug 1712: Introduce $prdr_requested flag variable
1139 JH/16 Bug 1714: Permit an empty string as expansion result for transport
1140 option transport_filter, meaning no filtering.
1142 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
1144 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
1145 defaults to "*" (all hosts). The variable is now available when not built
1146 with TLS, default unset, mainly to enable keeping the testsuite sane.
1147 If a server certificate is not supplied (via tls_certificate) an error is
1148 logged, and clients will find TLS connections fail on startup. Presumably
1149 they will retry in-clear.
1150 Packagers of Exim are strongly encouraged to create a server certificate
1151 at installation time.
1153 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
1154 with the $config_file variable.
1156 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
1157 in transport context, after the attempt, and per-recipient. The latter type
1158 is per host attempted. The event data is the error message, and the errno
1159 information encodes the lookup type (A vs. MX) used for the (first) host,
1160 and the trailing two digits of the smtp 4xx response.
1162 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
1163 to write to mainlog (or rejectlog, paniclog) in the window between file
1164 creation and permissions/ownership being changed. Particularly affects
1165 installations where exicyclog is run as root, rather than exim user;
1166 result is that the running daemon panics and dies.
1168 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
1170 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
1171 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
1172 "pri" and "weight". Note that the previous implicit priority given by the
1173 list order is no longer honoured.
1175 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
1176 for DKIM processing.
1178 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
1179 by defining SUPPORT_SOCKS.
1181 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
1182 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
1183 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
1184 variables are renamed to proxy_{local,external}_{address,port}.
1186 JH/27 Move Internationalisation support from Experimental to mainline, enabled
1187 for a build by defining SUPPORT_I18N
1189 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
1190 of the query string, and make ${quote_redis:} do that quoting.
1192 JH/29 Move Events support from Experimental to mainline, enabled by default
1193 and removable for a build by defining DISABLE_EVENT.
1195 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
1197 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
1198 cached by the daemon.
1200 JH/32 Move Redis support from Experimental to mainline, enabled for a build
1201 by defining LOOKUP_REDIS. The libhiredis library is required.
1203 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
1204 keys are given for lookup.
1206 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
1207 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
1208 only supported when built with TLS support. The PolarSSL SHA routines
1209 are still used when the TLS library is too old for convenient support.
1211 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
1212 openssl_options), for security. OpenSSL forces this from version 1.1.0
1213 server-side so match that on older versions.
1215 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
1216 allocation for $value could be released as the expansion processing
1217 concluded, but leaving the global pointer active for it.
1219 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
1220 and to use the domains and local_parts ACL conditions.
1222 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
1223 incorrectly not doubled on cutthrough transmission, hence seen as a
1224 body-termination at the receiving system - resulting in truncated mails.
1225 Commonly the sender saw a TCP-level error, and retransmitted the message
1226 via the normal store-and-forward channel. This could result in duplicates
1227 received - but deduplicating mailstores were liable to retain only the
1228 initial truncated version.
1230 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
1232 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
1234 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
1235 we're in there, support oversigning also; bug 1309.
1237 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
1239 HS/04 Add support for keep_environment and add_environment options.
1241 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
1242 either intentional arithmetic overflow during PRNG, or testing config-
1245 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
1246 delivery resulted in actual delivery. Cancel cutthrough before DATA
1249 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
1250 rejects a recipient: pass the reject to the originator.
1252 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
1253 Many were false-positives and ignorable, but it's worth fixing the
1256 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
1257 for the new environment-manipulation done at startup. Move the routines
1258 from being local to tls.c to being global via the os.c file.
1260 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
1261 an extract embedded as result-arg for a map, the first arg for extract
1262 is unavailable so we cannot tell if this is a numbered or keyed
1263 extraction. Accept either.
1269 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
1272 JH/02 The smtp transport option "multi_domain" is now expanded.
1274 JH/03 The smtp transport now requests PRDR by default, if the server offers
1277 JH/04 Certificate name checking on server certificates, when exim is a client,
1278 is now done by default. The transport option tls_verify_cert_hostnames
1279 can be used to disable this per-host. The build option
1280 EXPERIMENTAL_CERTNAMES is withdrawn.
1282 JH/05 The value of the tls_verify_certificates smtp transport and main options
1283 default to the word "system" to access the system default CA bundle.
1284 For GnuTLS, only version 3.0.20 or later.
1286 JH/06 Verification of the server certificate for a TLS connection is now tried
1287 (but not required) by default. The verification status is now logged by
1288 default, for both outbound TLS and client-certificate supplying inbound
1291 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
1294 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
1295 Status Notification (bounce) messages are now MIME format per RFC 3464.
1296 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
1297 under the control of the dsn_advertise_hosts option, and routers may
1298 have a dsn_lasthop option.
1300 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
1301 default, modifiable by a malware= option. The list separator for
1302 the options can now be changed in the usual way. Bug 68.
1304 JH/10 The smtp_receive_timeout main option is now expanded before use.
1306 JH/11 The incoming_interface log option now also enables logging of the
1307 local interface on delivery outgoing connections.
1309 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
1310 if the interface and destination host and port all match.
1312 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
1315 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
1316 Patch from Andrew Lewis.
1318 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
1319 now supports optional time-restrictions, weighting, and priority
1320 modifiers per server. Patch originally by <rommer@active.by>.
1322 JH/16 The spamd_address main option now supports a mixed list of local
1323 and remote servers. Remote servers can be IPv6 addresses, and
1324 specify a port-range.
1326 JH/17 Bug 68: The spamd_address main option now supports an optional
1327 timeout value per server.
1329 JH/18 Bug 1581: Router and transport options headers_add/remove can
1330 now have the list separator specified.
1332 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
1335 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
1338 JH/21 Support for the A6 type of dns record is withdrawn.
1340 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
1341 rather than the verbs used.
1343 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
1344 from 255 to 1024 chars.
1346 JH/24 Verification callouts now attempt to use TLS by default.
1348 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
1349 are generic router options now. The defaults didn't change.
1351 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
1352 Original patch from Alexander Shikoff, worked over by JH.
1354 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
1355 files if ZCAT_COMMAND is not executable.
1357 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
1359 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
1361 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
1362 Normally benign, it bites when the pair was led to by a CNAME;
1363 modern usage is to not canonicalize the domain to a CNAME target
1364 (and we were inconsistent anyway for A-only vs AAAA+A).
1366 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
1368 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
1369 when evaluating $sender_host_dnssec.
1371 JH/31 Check the HELO verification lookup for DNSSEC, adding new
1372 $sender_helo_dnssec variable.
1374 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
1376 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
1378 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
1380 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
1381 documented as working, but never had. Support all but $spam_report.
1383 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
1384 added for tls authenticator.
1386 HS/03 Add perl_taintmode main config option
1392 TL/01 When running the test suite, the README says that variables such as
1393 no_msglog_check are global and can be placed anywhere in a specific
1394 test's script, however it was observed that placement needed to be near
1395 the beginning for it to behave that way. Changed the runtest perl
1396 script to read through the entire script once to detect and set these
1397 variables, reset to the beginning of the script, and then run through
1398 the script parsing/test process like normal.
1400 TL/02 The BSD's have an arc4random API. One of the functions to induce
1401 adding randomness was arc4random_stir(), but it has been removed in
1402 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
1403 function when detected.
1405 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
1406 cause callback expansion.
1408 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
1409 syntax errors in an expansion can be treated as a string instead of
1410 logging or causing an error, due to the internal use of bool_lax
1411 instead of bool when processing it.
1413 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
1414 server certificates when making smtp deliveries.
1416 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
1418 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
1420 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
1422 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
1423 Merged patch from Sebastian Wiedenroth.
1425 JH/05 Fix results-pipe from transport process. Several recipients, combined
1426 with certificate use, exposed issues where response data items split
1427 over buffer boundaries were not parsed properly. This eventually
1428 resulted in duplicates being sent. This issue only became common enough
1429 to notice due to the introduction of connection certificate information,
1430 the item size being so much larger. Found and fixed by Wolfgang Breyha.
1432 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
1433 size buffer was used, resulting in syntax errors when an expansion
1436 JH/07 Add support for directories of certificates when compiled with a GnuTLS
1437 version 3.3.6 or later.
1439 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
1440 is EXPERIMENTAL_EVENT, the main-configuration and transport options
1441 both become "event_action", the variables become $event_name, $event_data
1442 and $event_defer_errno. There is a new variable $verify_mode, usable in
1443 routers, transports and related events. The tls:cert event is now also
1444 raised for inbound connections, if the main configuration event_action
1447 TL/06 In test suite, disable OCSP for old versions of openssl which contained
1448 early OCSP support, but no stapling (appears to be less than 1.0.0).
1450 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
1451 server certificate names available under the smtp transport option
1452 "tls_verify_cert_hostname" now do not permit multi-component wildcard
1455 JH/10 Time-related extraction expansions from certificates now use the main
1456 option "timezone" setting for output formatting, and are consistent
1457 between OpenSSL and GnuTLS compilations. Bug 1541.
1459 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
1460 encoded parameter in the incoming message. Bug 1558.
1462 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
1463 include certificate info, eximon was claiming there were spoolfile
1466 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
1468 JH/14 Log delivery-related information more consistently, using the sequence
1469 "H=<name> [<ip>]" wherever possible.
1471 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
1472 are problematic for Debian distribution, omit them from the release
1475 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
1477 JH/16 Fix string representation of time values on 64bit time_t architectures.
1480 JH/17 Fix a null-indirection in certextract expansions when a nondefault
1481 output list separator was used.
1486 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
1487 checkers that were complaining about end of non-void function with no
1490 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
1491 This was a regression introduced in 4.83 by another bugfix.
1493 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1495 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1496 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1502 TF/01 Correctly close the server side of TLS when forking for delivery.
1504 When a message was received over SMTP with TLS, Exim failed to clear up
1505 the incoming connection properly after forking off the child process to
1506 deliver the message. In some situations the subsequent outgoing
1507 delivery connection happened to have the same fd number as the incoming
1508 connection previously had. Exim would try to use TLS and fail, logging
1509 a "Bad file descriptor" error.
1511 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1512 utilities have not been installed.
1514 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1515 temporary space as the ACL may create new global variables.
1517 TL/01 LDAP support uses per connection or global context settings, depending
1518 upon the detected version of the libraries at build time.
1520 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1521 to extract and use the src ip:port in logging and expansions as if it
1522 were a direct connection from the outside internet. PPv2 support was
1523 updated based on HAProxy spec change in May 2014.
1525 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1527 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1528 Properly escape header and check for NULL return.
1530 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1533 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1535 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1536 characters in header names, implemented as a verify condition.
1537 Contributed by Michael Fischer v. Mollard.
1539 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1540 results permerror and temperror. Previous values are deprecated but
1541 still accepted. In a future release, err_perm and err_temp will be
1542 completely removed, which will be a backward incompatibility if the
1543 ACL tests for either of these two old results. Patch contributed by
1544 user bes-internal on the mailing list.
1546 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1548 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1549 selectors, in both main and reject logs.
1551 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1554 JH/07 Add malware type "sock" for talking to simple daemon.
1556 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1558 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1559 routers/transports under cutthrough routing.
1561 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1562 numbers. Touch up "bool" conditional to keep the same definition.
1564 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1566 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1568 JH/12 Expand items in router/transport headers_add or headers_remove lists
1569 individually rather than the list as a whole. Bug 1452.
1571 Required for reasonable handling of multiple headers_ options when
1572 they may be empty; requires that headers_remove items with embedded
1573 colons must have them doubled (or the list-separator changed).
1575 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1576 view the policy declared in the DMARC record. Currently, $dmarc_status
1577 is a combined value of both the record presence and the result of the
1580 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1582 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1583 dnslookup router and the smtp transport (applying to the forward
1586 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1587 of ldap servers used for a specific lookup. Patch provided by Heiko
1590 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1591 New variable $lookup_dnssec_authenticated for observability.
1593 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1594 Patch submitted by Lars Timman.
1596 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1598 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1599 Requires trusted mode and valid format message id, aborts otherwise.
1600 Patch contributed by Heiko Schlichting.
1602 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1603 certextract with support for various fields. Bug 1358.
1605 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1606 is requested by default, modifiable by smtp transport option
1609 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1610 operate on certificate variables to give certificate fingerprints
1611 Also new ${sha256:cert_variable}.
1613 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1615 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1616 Christian Aistleitner.
1618 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1620 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1621 file. Patch from Wolfgang Breyha.
1623 JH/25 Expand the coverage of the delivery $host and $host_address to
1624 client authenticators run in verify callout. Bug 1476.
1626 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1627 align with daemon_smtp_ports. Bug 72.
1629 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1630 support and error reporting did not work properly.
1632 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1633 and is readable. Patch from Andrew Colin Kissa.
1635 TL/14 Enhance documentation of ${run expansion and how it parses the
1636 commandline after expansion, particularly in the case when an
1637 unquoted variable expansion results in an empty value.
1639 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1641 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1642 by renaming away from it. Addresses GH issue 15, reported by
1645 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1647 TL/15 SECURITY: prevent double expansion in math comparison functions
1648 (can expand unsanitized data). Not remotely exploitable.
1655 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1657 PP/02 Make -n do something, by making it not do something.
1658 When combined with -bP, the name of an option is not output.
1660 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1663 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1664 $sender_host_name and config options to manage this, and basic check
1667 PP/05 DSCP support for outbound connections and control modifier for inbound.
1669 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1670 (Only plugin which currently uses this is kerberos4, which nobody should
1671 be using, but we should make it available and other future plugins might
1672 conceivably use it, even though it would break NAT; stuff *should* be
1673 using channel bindings instead).
1675 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1676 name; added for Sendmail compatibility; requires admin caller.
1677 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1678 just ignore it); requires trusted caller.
1679 Also parse but ignore: -Ac -Am -X<logfile>
1682 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1684 TL/02 Add +smtp_confirmation as a default logging option.
1686 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1687 Patch by Magnus Holmgren from 2007-02-20.
1689 TL/04 Bugzilla 1281 - Spec typo.
1690 Bugzilla 1283 - Spec typo.
1691 Bugzilla 1290 - Spec grammar fixes.
1693 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1695 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1697 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1698 mailing list by by Dmitry Isaikin.
1700 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1702 JH/02 Support "G" suffix to numbers in ${if comparisons.
1704 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1706 NM/01 Bugzilla 1197 - Spec typo
1707 Bugzilla 1196 - Spec examples corrections
1709 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1711 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1712 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1715 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1716 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1718 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1719 "acl {{name}{arg}...}", and optional args on acl condition
1722 JH/05 Permit multiple router/transport headers_add/remove lines.
1724 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1726 JH/07 Avoid using a waiting database for a single-message-only transport.
1727 Performance patch from Paul Fisher. Bugzilla 1262.
1729 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1732 JH/09 Add $headers_added variable, with content from use of ACL modifier
1733 add_header (but not yet added to the message). Bugzilla 199.
1735 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1736 Pulled from Bugzilla 817 by Wolfgang Breyha.
1738 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1740 (nb: this is the same fix as in Exim 4.80.1)
1742 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1745 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1748 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1750 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1751 advertise SMTP AUTH mechanism to us, instead of a generic
1752 protocol violation error. Also, make Exim more robust to bad
1753 data from the Dovecot auth socket.
1755 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1757 When a queue runner is handling a message, Exim first routes the
1758 recipient addresses, during which it prunes them based on the retry
1759 hints database. After that it attempts to deliver the message to
1760 any remaining recipients. It then updates the hints database using
1763 So if a recipient address works intermittently, it can get repeatedly
1764 deferred at routing time. The retry hints record remains fresh so the
1765 address never reaches the final cutoff time.
1767 This is a fairly common occurrence when a user is bumping up against
1768 their storage quota. Exim had some logic in its local delivery code
1769 to deal with this. However it did not apply to per-recipient defers
1770 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1772 This change adds a proper retry rule check during routing so that the
1773 final cutoff time is checked against the message's age. We only do
1774 this check if there is an address retry record and there is not a
1775 domain retry record; this implies that previous attempts to handle
1776 the address had the retry_use_local_parts option turned on. We use
1777 this as an approximation for the destination being like a local
1778 delivery, as in LMTP.
1780 I suspect this new check makes the old local delivery cutoff check
1781 redundant, but I have not verified this so I left the code in place.
1783 TF/02 Correct gecos expansion when From: is a prefix of the username.
1785 Test 0254 submits a message to Exim with the header
1789 When I ran the test suite under the user fanf2, Exim expanded
1790 the header to contain my full name, whereas it should have added
1791 a Resent-Sender: header. It erroneously treats any prefix of the
1792 username as equal to the username.
1794 This change corrects that bug.
1796 GF/01 DCC debug and logging tidyup
1797 Error conditions log to paniclog rather than rejectlog.
1798 Debug lines prefixed by "DCC: " to remove any ambiguity.
1800 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1802 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1803 Bug spotted by Jeremy Harris; was flawed since initial commit.
1804 Would have resulted in OCSP responses post-SNI triggering an Exim
1805 NULL dereference and crash.
1807 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1809 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1810 Bug detection, analysis and fix by Samuel Thibault.
1811 Bugzilla 1331, Debian bug #698092.
1813 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1815 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1816 Server implementation by Todd Lyons, client by JH.
1817 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1818 config variable "prdr_enable" controls whether the server
1819 advertises the facility. If the client requests PRDR a new
1820 acl_data_smtp_prdr ACL is called once for each recipient, after
1821 the body content is received and before the acl_smtp_data ACL.
1822 The client is controlled by both of: a hosts_try_prdr option
1823 on the smtp transport, and the server advertisement.
1824 Default client logging of deliveries and rejections involving
1825 PRDR are flagged with the string "PRDR".
1827 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1828 fclose(). Diagnosis by Todd Lyons.
1830 PP/17 Update configure.default to handle IPv6 localhost better.
1831 Patch by Alain Williams (plus minor tweaks).
1834 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1835 This is now consistent with GnuTLS, and is now documented: the
1836 previous undocumented portable approach to treating the option as
1837 unset was to force an expansion failure. That still works, and
1838 an empty string is now equivalent.
1840 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1841 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1842 not performing validation itself.
1844 PP/20 Added force_command boolean option to pipe transport.
1845 Patch from Nick Koston, of cPanel Inc.
1847 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1850 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1852 PP/21 Fix eximon continuous updating with timestamped log-files.
1853 Broken in a format-string cleanup in 4.80, missed when I repaired the
1854 other false fix of the same issue.
1855 Report and fix from Heiko Schlichting.
1858 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1859 Report from Prashanth Katuri.
1861 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1862 It's SecureTransport, so affects any MacOS clients which use the
1863 system-integrated TLS libraries, including email clients.
1865 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1866 using a MIME ACL for non-SMTP local injection.
1867 Report and assistance in diagnosis by Warren Baker.
1869 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1871 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1873 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1874 last id that failed so it may be referenced in subsequent ACL's.
1876 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1879 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1880 ldap library initialization, allowing self-signed CA's to be
1881 used. Also properly sets require_cert option later in code by
1882 using NULL (global ldap config) instead of ldap handle (per
1883 session). Bug diagnosis and testing by alxgomz.
1885 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1886 the src/util/ subdirectory.
1888 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1889 renamed to Transport Post Delivery Action by Jeremy Harris, as
1892 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1893 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1894 redis_servers = needs to be configured which will be used by the redis
1895 lookup. Patch from Warren Baker, of The Packet Hub.
1897 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1899 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1900 hostname or reverse DNS when processing a host list. Used suggestions
1901 from multiple comments on this bug.
1903 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1905 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1906 Missed a few lines, added it to make the runtest require no keyboard
1909 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1910 contains upper case chars. Make router use caseful_local_part.
1912 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1913 support when GnuTLS has been built with p11-kit.
1919 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1921 This, or similar/improved, will also be change PP/11 of 4.82.
1927 PP/01 Handle short writes when writing local log-files.
1928 In practice, only affects FreeBSD (8 onwards).
1929 Bugzilla 1053, with thanks to Dmitry Isaikin.
1931 NM/01 Bugzilla 949 - Documentation tweak
1933 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1936 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1938 PP/02 Implemented gsasl authenticator.
1940 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1942 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1943 `pkg-config foo` for cflags/libs.
1945 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1946 with rest of GSASL and with heimdal_gssapi.
1948 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1949 `pkg-config foo` for cflags/libs for the TLS implementation.
1951 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1952 properties get this fed in as external SSF. A number of robustness
1953 and debugging improvements to the cyrus_sasl authenticator.
1955 PP/08 cyrus_sasl server now expands the server_realm option.
1957 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1958 Patch by Jeremy Harris.
1960 PP/10 Added dbmjz lookup type.
1962 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1964 PP/12 MAIL args handles TAB as well as SP, for better interop with
1965 non-compliant senders.
1966 Analysis and variant patch by Todd Lyons.
1968 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1969 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1970 Patch from Dirk Mueller <dmueller@suse.com>
1972 PP/13 tls_peerdn now print-escaped for spool files.
1973 Observed some $tls_peerdn in wild which contained \n, which resulted
1974 in spool file corruption.
1976 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1977 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1978 or write after TLS renegotiation, which otherwise led to messages
1981 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1982 as a tracking header (ie: a signed header comes before the signature).
1983 Patch from Wolfgang Breyha.
1985 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1986 comma-sep list; embedded commas doubled.
1988 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1990 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1992 Report and patch from Dmitry Banschikov.
1994 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1995 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1996 failures. We appear to now support TLS1.1+ with Exim.
1998 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1999 lets Exim select keys and certificates based upon TLS SNI from client.
2000 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
2001 before an outbound SMTP session. New log_selector, +tls_sni.
2003 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
2004 NULL dereference. Report and patch from Alun Jones.
2006 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
2007 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
2008 Not seeing resolver debug output on NetBSD, but suspect this is a
2009 resolver implementation change.
2011 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
2012 Left warnings. Added "eximon gdb" invocation mode.
2014 PP/21 Defaulting "accept_8bitmime" to true, not false.
2016 PP/22 Added -bw for inetd wait mode support.
2018 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
2019 locate the relevant includes and libraries. Made this the default.
2021 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
2022 Bugzilla 1246, report and most of solution from Tomasz Kusy.
2024 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
2025 This may cause build issues on older platforms.
2027 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
2028 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
2029 gnutls_require_mac & gnutls_require_protocols (no longer supported).
2030 Added SNI support via GnuTLS too.
2031 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
2033 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
2035 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
2036 Applied second patch from Janne, implementing suggestion to default
2037 multiple-strings-in-record handling to match SPF spec.
2039 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
2041 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
2042 read-only, out of scope).
2043 Patch from Wolfgang Breyha, report from Stuart Northfield.
2045 PP/29 Fix three issues highlighted by clang analyser static analysis.
2046 Only crash-plausible issue would require the Cambridge-specific
2047 iplookup router and a misconfiguration.
2048 Report from Marcin Mirosław.
2050 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
2052 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
2053 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
2054 As part of this, removing so much warning spew let me fix some minor
2055 real issues in debug logging.
2057 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
2058 assignment on my part. Fixed.
2060 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
2061 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
2062 Janne Snabb (who went above and beyond: thank you).
2064 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
2065 string otherwise requires a connection and a bunch more work and it's
2066 relatively easy to get wrong. Should also expose TLS library linkage
2069 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
2070 64-bit ${eval} (JH/03).
2072 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
2073 GNU libc to support some of the 64-bit stuff, should not lead to
2074 conflicts. Defined before os.h is pulled in, so if a given platform
2075 needs to override this, it can.
2077 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
2078 protection layer was required, which is not implemented.
2079 Bugzilla 1254, patch from Wolfgang Breyha.
2081 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
2082 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
2083 tls_dhparam take prime identifiers. Also unbreak combination of
2084 OpenSSL+DH_params+TLSSNI.
2086 PP/39 Disable SSLv2 by default in OpenSSL support.
2092 PP/01 Solaris build fix for Oracle's LDAP libraries.
2093 Bugzilla 1109, patch from Stephen Usher.
2095 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
2097 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
2100 TF/02 Fix a couple more cases where we did not log the error message
2101 when unlink() failed. See also change 4.74-TF/03.
2103 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
2104 lock up or crash if it happened to be inside a call to libc when it
2105 got a SIGUSR1 from exiwhat.
2107 The SIGUSR1 handler appends the current process status to the process
2108 log which is later printed by exiwhat. It used to use the general
2109 purpose logging code to do this, but several functions it calls are
2110 not safe for signals.
2112 The new output code in the SIGUSR1 handler is specific to the process
2113 log, and simple enough that it's easy to inspect for signal safety.
2114 Removing some special cases also simplifies the general logging code.
2115 Removing the spurious timestamps from the process log simplifies
2118 TF/04 Improved ratelimit ACL condition.
2120 The /noupdate option has been deprecated in favour of /readonly which
2121 has clearer semantics. The /leaky, /strict, and /readonly update modes
2122 are mutually exclusive. The update mode is no longer included in the
2123 database key; it just determines when the database is updated. (This
2124 means that when you upgrade Exim will forget old rate measurements.)
2126 Exim now checks that the per_* options are used with an update mode that
2127 makes sense for the current ACL. For example, when Exim is processing a
2128 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
2129 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
2130 must specify per_mail/readonly. If you omit the update mode it defaults to
2131 /leaky where that makes sense (as before) or /readonly where required.
2133 The /noupdate option is now undocumented but still supported for
2134 backwards compatibility. It is equivalent to /readonly except that in
2135 ACLs where /readonly is required you may specify /leaky/noupdate or
2136 /strict/noupdate which are treated the same as /readonly.
2138 A useful new feature is the /count= option. This is a generalization
2139 of the per_byte option, so that you can measure the throughput of other
2140 aggregate values. For example, the per_byte option is now equivalent
2141 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
2143 The per_rcpt option has been generalized using the /count= mechanism
2144 (though it's more complicated than the per_byte equivalence). When it is
2145 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
2146 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
2147 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
2148 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
2149 Note that using per_rcpt with a non-readonly update mode in more than
2150 one ACL will cause the recipients to be double-counted. (The per_mail
2151 and per_byte options don't have this problem.)
2153 The handling of very low rates has changed slightly. If the computed rate
2154 is less than the event's count (usually one) then this event is the first
2155 after a long gap. In this case the rate is set to the same as this event's
2156 count, so that the first message of a spam run is counted properly.
2158 The major new feature is a mechanism for counting the rate of unique
2159 events. The new per_addr option counts the number of different
2160 recipients that someone has sent messages to in the last time period. It
2161 behaves like per_rcpt if all the recipient addresses are different, but
2162 duplicate recipient addresses do not increase the measured rate. Like
2163 the /count= option this is a general mechanism, so the per_addr option
2164 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
2165 example, measure the rate that a client uses different sender addresses
2166 with the options per_mail/unique=$sender_address. There are further
2167 details in the main documentation.
2169 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
2171 TF/06 Removed a few PCRE remnants.
2173 TF/07 Automatically extract Exim's version number from tags in the git
2174 repository when doing development or release builds.
2176 PP/02 Raise smtp_cmd_buffer_size to 16kB.
2177 Bugzilla 879. Patch from Paul Fisher.
2179 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
2180 Heavily based on revision 40f9a89a from Simon Arlott's tree.
2183 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
2185 PP/05 Variable $av_failed, true if the AV scanner deferred.
2186 Bugzilla 1078. Patch from John Horne.
2188 PP/06 Stop make process more reliably on build failure.
2189 Bugzilla 1087. Patch from Heiko Schlittermann.
2191 PP/07 Make maildir_use_size_file an _expandable_ boolean.
2192 Bugzilla 1089. Patch from Heiko Schlittermann.
2194 PP/08 Handle ${run} returning more data than OS pipe buffer size.
2195 Bugzilla 1131. Patch from Holger Weiß.
2197 PP/09 Handle IPv6 addresses with SPF.
2198 Bugzilla 860. Patch from Wolfgang Breyha.
2200 PP/10 GnuTLS: support TLS 1.2 & 1.1.
2202 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
2205 PP/11 match_* no longer expand right-hand-side by default.
2206 New compile-time build option, EXPAND_LISTMATCH_RHS.
2207 New expansion conditions, "inlist", "inlisti".
2209 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
2211 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
2213 PP/14 fix log_write() format string regression from TF/03.
2214 Bugzilla 1152. Patch from Dmitry Isaikin.
2220 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
2222 PP/02 Harmonised TLS library version reporting; only show if debugging.
2223 Layout now matches that introduced for other libraries in 4.74 PP/03.
2225 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
2227 PP/04 New "dns_use_edns0" global option.
2229 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
2232 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
2233 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
2235 TK/01 Updated PolarSSL code to 0.14.2.
2236 Bugzilla 1097. Patch from Andreas Metzler.
2238 PP/07 Catch divide-by-zero in ${eval:...}.
2239 Fixes bugzilla 1102.
2241 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
2244 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
2245 format-string attack -- SECURITY: remote arbitrary code execution.
2247 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
2248 time unintentionally subject to list matching rules, letting the header
2249 cause arbitrary Exim lookups (of items which can occur in lists, *not*
2250 arbitrary string expansion). This allowed for information disclosure.
2252 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
2253 INT_MIN/-1 -- value coerced to INT_MAX.
2259 NM/01 Workaround for PCRE version dependency in version reporting
2262 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
2263 This fixes portability to compilers other than gcc, notably
2264 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
2266 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
2267 makefiles for portability to HP-UX and POSIX correctness.
2269 PP/01 Permit LOOKUP_foo enabling on the make command-line.
2270 Also via indented variable definition in the Makefile.
2271 (Debugging by Oliver Heesakkers).
2273 PP/02 Restore caching of spamd results with expanded spamd_address.
2274 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
2276 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
2277 Improves build reliability. Fix from: Frank Elsner
2279 NM/02 Fix wide character breakage in the rfc2047 coding
2280 Fixes bug 1064. Patch from Andrey N. Oktyabrski
2282 NM/03 Allow underscore in dnslist lookups
2283 Fixes bug 1026. Patch from Graeme Fowler
2285 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
2286 Code patches from Adam Ciarcinski of NetBSD.
2288 NM/04 Fixed exiqgrep to cope with mailq missing size issue
2291 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
2292 is logged, to avoid truncation. Patch from John Horne.
2294 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
2295 Patch from Jakob Hirsch.
2297 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
2298 SQL string expansion failure details.
2299 Patch from Andrey Oktyabrski.
2301 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
2302 Patch from Simon Arlott.
2304 PP/09 New lookups functionality failed to compile on old gcc which rejects
2305 extern declarations in function scope.
2306 Patch from Oliver Fleischmann
2308 PP/10 Use sig_atomic_t for flags set from signal handlers.
2309 Check getgroups() return and improve debugging.
2310 Fixed developed for diagnosis in bug 927 (which turned out to be
2313 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
2314 Patch from Mark Zealey.
2316 PP/12 Bugzilla 1056: Improved spamd server selection.
2317 Patch from Mark Zealey.
2319 PP/13 Bugzilla 1086: Deal with maildir quota file races.
2320 Based on patch from Heiko Schlittermann.
2322 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
2323 Patch from Uwe Doering, sign-off by Michael Haardt.
2325 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
2326 variable declaration deep within a block. Bug and patch from
2329 PP/15 lookups-Makefile IRIX compatibility coercion.
2331 PP/16 Make DISABLE_DKIM build knob functional.
2333 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
2334 Patch by Simon Arlott
2336 TF/03 Fix valgrind.h portability to C89 compilers that do not support
2337 variable argument macros. Our copy now differs from upstream.
2343 TF/01 Failure to get a lock on a hints database can have serious
2344 consequences so log it to the panic log.
2346 TF/02 Log LMTP confirmation messages in the same way as SMTP,
2347 controlled using the smtp_confirmation log selector.
2349 TF/03 Include the error message when we fail to unlink a spool file.
2351 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
2352 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
2353 for maintaining out-of-tree patches for some time.
2355 PP/01 Bugzilla 139: Documentation and portability issues.
2356 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
2357 Handle per-OS dynamic-module compilation flags.
2359 PP/02 Let /dev/null have normal permissions.
2360 The 4.73 fixes were a little too stringent and complained about the
2361 permissions on /dev/null. Exempt it from some checks.
2362 Reported by Andreas M. Kirchwitz.
2364 PP/03 Report version information for many libraries, including
2365 Exim version information for dynamically loaded libraries. Created
2366 version.h, now support a version extension string for distributors
2367 who patch heavily. Dynamic module ABI change.
2369 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
2370 privilege escalation vulnerability whereby the Exim run-time user
2371 can cause root to append content of the attacker's choosing to
2374 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
2377 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
2378 If dropping privileges for untrusted macros, we disabled normal logging
2379 on the basis that it would fail; for the Exim run-time user, this is not
2380 the case, and it resulted in successful deliveries going unlogged.
2381 Fixed. Reported by Andreas Metzler.
2387 PP/01 Date: & Message-Id: revert to normally being appended to a message,
2388 only prepend for the Resent-* case. Fixes regression introduced in
2389 Exim 4.70 by NM/22 for Bugzilla 607.
2391 PP/02 Include check_rfc2047_length in configure.default because we're seeing
2392 increasing numbers of administrators be bitten by this.
2394 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
2396 PP/03 Bugzilla 994: added openssl_options main configuration option.
2398 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
2400 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
2402 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
2404 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
2405 without a peer certificate, leading to a segfault because of an
2406 assumption that peers always have certificates. Be a little more
2407 paranoid. Problem reported by Martin Tscholak.
2409 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
2410 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
2411 NB: ClamAV planning to remove STREAM in "middle of 2010".
2412 CL also introduces -bmalware, various -d+acl logging additions and
2413 more caution in buffer sizes.
2415 PP/09 Implemented reverse_ip expansion operator.
2417 PP/10 Bugzilla 937: provide a "debug" ACL control.
2419 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
2421 PP/12 Bugzilla 973: Implement --version.
2423 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
2425 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
2427 PP/15 Bugzilla 816: support multiple condition rules on Routers.
2429 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
2430 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
2431 ignore trailing whitespace.
2433 JJ/02 prevent non-panic DKIM error from being sent to paniclog
2435 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
2438 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
2439 Notification from Dr Andrew Aitchison.
2441 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
2442 ExtendedDetectionInfo response format.
2443 Notification from John Horne.
2445 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
2448 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
2449 XSL and documented dependency on system catalogs, with examples of how
2452 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
2455 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
2456 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
2457 configuration file which is writeable by the Exim user or group.
2459 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
2460 of configuration files to cover files specified with the -C option if
2461 they are going to be used with root privileges, not just the default
2464 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
2465 option (effectively making it always true).
2467 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
2468 files to be used while preserving root privileges.
2470 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
2471 that rogue child processes cannot use them.
2473 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
2474 run-time user, instead of root.
2476 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
2477 Exim run-time user without dropping privileges.
2479 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
2480 result string, instead of calling string_vformat() twice with the same
2483 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
2484 for other users. Others should always drop root privileges if they use
2485 -C on the command line, even for a whitelisted configure file.
2487 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
2489 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2495 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2496 $data_path, and $header_path variables; fixed documentation bugs and
2499 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2500 exipick to access non-standard spools, including the "frozen" queue
2503 NM/01 Bugzilla 965: Support mysql stored procedures.
2504 Patch from Alain Williams
2506 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2508 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2509 Patch from Andreas Metzler
2511 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2512 Patch from Kirill Miazine
2514 NM/05 Bugzilla 671: Added umask to procmail example.
2516 JJ/03 installed exipick 20100323.0, fixing doc bug
2518 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2519 directory. Notification and patch from Dan Rosenberg.
2521 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2523 TK/02 Improve log output when DKIM signing operation fails.
2525 MH/01 Treat the transport option dkim_domain as a colon separated
2526 list, not as a single string, and sign the message with each element,
2527 omitting multiple occurences of the same signer.
2529 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2530 Bugzilla 985, 986. Patch by Simon Arlott
2532 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2533 Patch by Simon Arlott
2535 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2536 MBX locking. Notification from Dan Rosenberg.
2542 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2544 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2546 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2548 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2550 NM/04 Bugzilla 915: Flag broken perl installation during build.
2556 TK/01 Added patch by Johannes Berg that expands the main option
2557 "spamd_address" if it starts with a dollar sign.
2559 TK/02 Write list of recipients to X-Envelope-Sender header when building
2560 the mbox-format spool file for content scanning (suggested by Jakob
2563 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2564 (http://www.dcc-servers.net/) support via dccifd. Activated by
2565 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2567 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2568 by Mark Daniel Reidel <mr@df.eu>.
2570 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2571 When building exim an external PCRE library is now needed -
2572 PCRE is a system library on the majority of modern systems.
2573 See entry on PCRE_LIBS in EDITME file.
2575 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2576 conversation. Added nologin parameter to request.
2577 Patch contributed by Kirill Miazine.
2579 TF/01 Do not log submission mode rewrites if they do not change the address.
2581 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2583 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2584 log files in place. Contributed by Roberto Lima.
2586 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2588 TF/03 Bugzilla 615: When checking the local_parts router precondition
2589 after a local_part_suffix or local_part_prefix option, Exim now
2590 does not use the address's named list lookup cache, since this
2591 contains cached lookups for the whole local part.
2593 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2594 Robert Millan. Documentation is in experimental-spec.txt.
2596 TF/04 Bugzilla 668: Fix parallel build (make -j).
2598 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2600 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2601 Patch provided by Jan Srzednicki.
2603 TF/05 Leading white space used to be stripped from $spam_report which
2604 wrecked the formatting. Now it is preserved.
2606 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2607 that they are available at delivery time.
2609 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2611 TF/08 TLS error reporting now respects the incoming_interface and
2612 incoming_port log selectors.
2614 TF/09 Produce a more useful error message if an SMTP transport's hosts
2615 setting expands to an empty string.
2617 NM/06 Bugzilla 744: EXPN did not work under TLS.
2618 Patch provided by Phil Pennock.
2620 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2621 Patch provided by Richard Godbee.
2623 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2624 acl_smtp_notquit, added index entry.
2626 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2627 Patch provided by Eugene Bujak.
2629 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2630 accept(). Patch provided by Maxim Dounin.
2632 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2633 Patch provided by Phil Pennock.
2635 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2637 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2638 Patch provided by Brad "anomie" Jorsch.
2640 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2641 Patch provided by Dean Brooks.
2643 TK/05 Add native DKIM support (does not depend on external libraries).
2645 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2646 Patch provided by Graeme Fowler.
2648 NM/16 Bugzilla 851: Documentation example syntax fix.
2650 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2652 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2655 NM/19 Bugzilla 745: TLS version reporting.
2656 Patch provided by Phil Pennock.
2658 NM/20 Bugzilla 167: bool: condition support.
2659 Patch provided by Phil Pennock.
2661 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2662 clients. Patch provided by Phil Pennock.
2664 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2665 Patch provided by Brad "anomie" Jorsch.
2667 NM/23 Bugzilla 687: Fix misparses in eximstats.
2668 Patch provided by Heiko Schlittermann.
2670 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2671 Patch provided by Heiko Schlittermann.
2673 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2674 plus update to original patch.
2676 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2678 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2679 Patch provided by David Brownlee.
2681 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2683 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2685 NM/30 Bugzilla 888: TLS documentation bugfixes.
2687 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2689 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2690 Unlike the original bugzilla I have changed all shell scripts in src tree.
2692 NM/33 Bugzilla 898: Transport filter timeout fix.
2693 Patch by Todd Rinaldo.
2695 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2696 Patch by Serge Demonchaux.
2698 NM/35 Bugzilla 39: Base64 decode bug fixes.
2699 Patch by Jakob Hirsch.
2701 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2703 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2705 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2707 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2713 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2714 ALT-N's libdkim that I have put here:
2715 http://duncanthrax.net/exim-experimental/
2717 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2718 were called 'true' and it seems that C99 defines that as a reserved
2719 keyword to be used with 'bool' variable types. That means you could
2720 not include C99-style headers which use bools without triggering
2721 build errors in sieve.c.
2723 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2724 as mailq or other aliases. Changed the --help handling significantly
2725 to do whats expected. exim_usage() emits usage/help information.
2727 SC/01 Added the -bylocaldomain option to eximstats.
2729 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2731 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2733 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2739 PH/01 Another patch from the Sieve maintainer.
2741 PH/02 When an IPv6 address is converted to a string for single-key lookup
2742 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2743 dots are used instead of colons so that keys in lsearch files need not
2744 contain colons. This was done some time before quoting was made available
2745 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2746 (notated using the quote facility) so as to distinguish them from IPv4
2747 keys. This meant that lookups for IP addresses in host lists did not work
2748 for iplsearch lookups.
2750 This has been fixed by arranging for IPv6 addresses to be expressed with
2751 colons if the lookup type is iplsearch. This is not incompatible, because
2752 previously such lookups could never work.
2754 The situation is now rather anomalous, since one *can* have colons in
2755 ordinary lsearch keys. However, making the change in all cases is
2756 incompatible and would probably break a number of configurations.
2758 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2761 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2762 conversion specification without a maximum field width, thereby enabling
2763 a rogue spamd server to cause a buffer overflow. While nobody in their
2764 right mind would setup Exim to query an untrusted spamd server, an
2765 attacker that gains access to a server running spamd could potentially
2766 exploit this vulnerability to run arbitrary code as the Exim user.
2768 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2769 $primary_hostname instead of what libspf2 thinks the hosts name is.
2771 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2772 a directory entry by the name of the lookup key. Previously, if a
2773 symlink pointed to a non-existing file or a file in a directory that
2774 Exim lacked permissions to read, a lookup for a key matching that
2775 symlink would fail. Now it is enough that a matching directory entry
2776 exists, symlink or not. (Bugzilla 503.)
2778 PH/03 The body_linecount and body_zerocount variables are now exported in the
2781 PH/04 Added the $dnslist_matched variable.
2783 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2784 This means they are set thereafter only if the connection becomes
2787 PH/06 Added the client_condition to authenticators so that some can be skipped
2788 by clients under certain conditions.
2790 PH/07 The error message for a badly-placed control=no_multiline_responses left
2791 "_responses" off the end of the name.
2793 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2795 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2796 (without spaces) instead of just copying the configuration text.
2798 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2800 PH/11 Added $max_received_linelength.
2802 PH/12 Added +ignore_defer and +include_defer to host lists.
2804 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2805 way in which PCRE > 7.0 is built.
2807 PH/14 Implemented queue_only_load_latch.
2809 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2810 MAIL command. The effect was to mangle the value on 64-bit systems.
2812 PH/16 Another patch from the Sieve maintainer.
2814 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2816 PH/18 If a system quota error occurred while trying to create the file for
2817 a maildir delivery, the message "Mailbox is full" was not appended to the
2818 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2819 only to a quota excession during the actual writing of the file.
2821 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2822 characters?) which causes problems in log lines. The DN values are now
2823 passed through string_printing() before being added to log lines.
2825 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2826 and InterBase are left for another time.)
2828 PH/21 Added message_body_newlines option.
2830 PH/22 Guard against possible overflow in moan_check_errorcopy().
2832 PH/23 POSIX allows open() to be a macro; guard against that.
2834 PH/24 If the recipient of an error message contained an @ in the local part
2835 (suitably quoted, of course), incorrect values were put in $domain and
2836 $local_part during the evaluation of errors_copy.
2842 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2843 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2846 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2847 issue a MAIL command.
2849 PH/02 In an ACL statement such as
2851 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2853 if a client was not listed at all, or was listed with a value other than
2854 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2855 the condition was not true (as it should be), so access was not denied.
2856 The bug was that the ! inversion was incorrectly passed on to the second
2857 item. This has been fixed.
2859 PH/03 Added additional dnslists conditions == and =& which are different from
2860 = and & when the dns lookup returns more than one IP address.
2862 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2863 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2865 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2866 FSYNC, which compiles an option called disable_fsync that allows for
2867 bypassing fsync(). The documentation is heavily laced with warnings.
2869 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2871 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2872 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2873 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2874 including adding "make clean"; (3) Added -fPIC when compiling the test
2875 dynamically loaded module, to get rid of a warning.
2877 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2878 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2879 = 0s. The bug is otherwise harmless.
2881 PH/07 There was a bug in the dovecot authenticator such that the value of
2882 $auth1 could be overwritten, and so not correctly preserved, after a
2883 successful authentication. This usually meant that the value preserved by
2884 the server_setid option was incorrect.
2886 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2888 PH/09 Installed PCRE release 7.0.
2890 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2891 run for batched SMTP input. It is now run at the start of every message
2892 in the batch. While fixing this I discovered that the process information
2893 (output by running exiwhat) was not always getting set for -bs and -bS
2894 input. This is fixed, and it now also says "batched" for BSMTP.
2896 PH/11 Added control=no_pipelining.
2898 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2899 patch, slightly modified), and move the expansion of helo_data till after
2900 the connection is made in the smtp transport (so it can use these
2903 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2905 PH/14 Added log_selector = +pid.
2907 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2909 PH/16 Add ${if forany and ${if forall.
2911 PH/17 Added dsn_from option to vary the From: line in DSNs.
2913 PH/18 Flush SMTP output before performing a callout, unless control =
2914 no_callout_flush is set.
2916 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2917 was true (the default) a successful delivery failed to delete the retry
2918 item, thus causing premature timeout of the address. The bug is now
2921 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2923 PH/21 Long custom messages for fakedefer and fakereject are now split up
2924 into multiline responses in the same way that messages for "deny" and
2925 other ACL rejections are.
2927 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2928 with slight modification.
2930 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2931 draft, changing the syntax and factoring some duplicate code".
2933 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2934 for deliveries of the second and subsequent messages over the same SMTP
2937 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2938 ${reduce, with only minor "tidies".
2940 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2942 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2943 expansion side effects.
2945 PH/27 When a message times out after an over-quota error from an Exim-imposed
2946 quota, the bounce message says "mailbox is full". This message was not
2947 being given when it was a system quota that was exceeded. It now should
2950 MH/03 Made $recipients available in local_scan(). local_scan() already has
2951 better access to the recipient list through recipients_list[], but
2952 $recipients can be useful in postmaster-provided expansion strings.
2954 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2955 in the case of a MAIL command with additional options following the
2956 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2957 were accidentally chopped off.
2959 PH/29 SMTP synchronization checks are implemented when a command is read -
2960 there is a check that no more input is waiting when there shouldn't be
2961 any. However, for some commands, a delay in an ACL can mean that it is
2962 some time before the response is written. In this time, more input might
2963 arrive, invalidly. So now there are extra checks after an ACL has run for
2964 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2965 pipelining has not been advertised.
2967 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2969 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2970 returned in the SMTP response when smtp_return_error_details was false.
2971 This has been fixed.
2973 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2974 instead of the C I/O that was originally supplied, because problems were
2975 reported on Solaris.
2977 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2978 Exim which did not show up earlier: it was assuming that a call to
2979 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2980 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2981 was a macro that became an assignment, so it seemed to work. This has
2982 changed to a proper function call with a void return, hence the compile
2983 error. Exim's code has been fixed.
2985 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2988 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2990 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2992 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2993 because it turns out to be too big or there is a timeout) there is no
2994 "Completed" line in the log. When some messages of this type were
2995 selected by exigrep, they were listed as "not completed". Others were
2996 picked up by some special patterns. I have improved the selection
2997 criteria to be more general.
2999 PH/38 The host_find_failed option in the manualroute router can now be set
3000 to "ignore", to completely ignore a host whose IP address cannot be
3001 found. If all hosts are ignored, the behaviour is controlled by the new
3002 host_all_ignored option.
3004 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
3005 homing or because of multiple MX records with /mx) generated more than
3006 one IP address, and the following item turned out to be the local host,
3007 all the secondary addresses of the first item were incorrectly removed
3008 from the list, along with the local host and any following hosts (which
3009 is what is supposed to happen).
3011 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
3012 whoever called Exim into the -H file. In the case of the daemon it was
3013 behaving confusingly. When first started, it used values for whoever
3014 started the daemon, but after a SIGHUP it used the Exim user (because it
3015 calls itself on a restart). I have changed the code so that it now always
3018 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
3019 message are rejected with the same error (e.g. no authentication or bad
3020 sender address), and a DATA command is nevertheless sent (as can happen
3021 with PIPELINING or a stupid MUA), the error message that was given to the
3022 RCPT commands is included in the rejection of the DATA command. This is
3023 intended to be helpful for MUAs that show only the final error to their
3026 PH/42 Another patch from the Sieve maintainer.
3028 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
3029 Eximstats - Fixed some broken HTML links and added missing column headers
3031 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
3032 columns for Rejects, Temp Rejects, Ham, and Spam rows.
3034 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
3036 PH/43 Yet another patch from the Sieve maintainer.
3038 PH/44 I found a way to check for a TCP/IP connection going away before sending
3039 the response to the final '.' that terminates a message, but only in the
3040 case where the client has not sent further data following the '.'
3041 (unfortunately, this is allowed). However, in many cases there won't be
3042 any further data because there won't be any more messages to send. A call
3043 to select() can be used: if it shows that the input is "ready", there is
3044 either input waiting, or the socket has been closed. An attempt to read
3045 the next input character can distinguish the two cases. Previously, Exim
3046 would have sent an OK response which the client would never have see.
3047 This could lead to message repetition. This fix should cure that, at
3048 least in a lot of common cases.
3050 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
3051 advertised in response to EHLO.
3057 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
3058 fixed by 4.65/MH/01 (is this a record?) are fixed:
3060 (i) An empty string was always treated as zero by the numeric comparison
3061 operators. This behaviour has been restored.
3063 (ii) It is documented that the numeric comparison operators always treat
3064 their arguments as decimal numbers. This was broken in that numbers
3065 starting with 0 were being interpreted as octal.
3067 While fixing these problems I realized that there was another issue that
3068 hadn't been noticed. Values of message_size_limit (both the global option
3069 and the transport option) were treated as octal if they started with 0.
3070 The documentation was vague. These values are now always treated as
3071 decimal, and I will make that clear in the documentation.
3077 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
3078 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
3081 MH/01 Don't check that the operands of numeric comparison operators are
3082 integers when their expansion is in "skipping" mode (fixes bug
3083 introduced by 4.64-PH/07).
3085 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
3086 child addresses, Exim now panics and dies. Previously, because the count
3087 is held in a short int, deliveries were likely to be lost. As such a
3088 large number of recipients for a single message is ridiculous
3089 (performance will be very, very poor), I have chosen to impose a limit
3090 rather than extend the field.
3096 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
3097 leftover -K file (the existence of which was triggered by #402).
3098 While we were at it, introduced process PID as part of the -K
3099 filename. This should rule out race conditions when creating
3102 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
3103 processing considerably. Previous code took too long for large mails,
3104 triggering a timeout which in turn triggers #401.
3106 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
3107 in the DK code in transports.c. sendfile() is not really portable,
3108 hence the _LINUX specificness.
3110 TF/01 In the add_headers option to the mail command in an Exim filter,
3111 there was a bug that Exim would claim a syntax error in any
3112 header after the first one which had an odd number of characters
3115 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
3116 callout verification, Exim cached a "reject" for the entire domain. This
3117 is correct for most verifications, but it is not correct for a recipient
3118 verification with use_sender or use_postmaster set, because in that case
3119 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
3120 case of MAIL FROM:<> rejection from other early rejections (e.g.
3121 rejection of HELO). When verifying a recipient using a non-null MAIL
3122 address, the cache is ignored if it shows MAIL FROM:<> rejection.
3123 Whatever the result of the callout, the value of the domain cache is
3124 left unchanged (for any other kind of callout, getting as far as trying
3125 RCPT means that the domain itself is ok).
3127 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
3130 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
3131 manifest itself as EPIPE rather than ECONNECT. When tidying away a
3132 session, the daemon ignores ECONNECT errors and logs others; it now
3133 ignores EPIPE as well.
3135 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
3136 (quoted-printable decoding).
3138 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
3139 later the small subsequent patch to fix an introduced bug.
3141 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
3143 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
3145 PH/08 An error is now given if message_size_limit is specified negative.
3147 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
3148 to be given (somewhat) arbitrary names.
3150 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
3153 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
3154 miscellaneous code fixes
3156 PH/10 Added the log_reject_target ACL modifier to specify where to log
3159 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
3160 hostname. This is wrong, because it relates to the incoming message (and
3161 probably the interface on which it is arriving) and not to the outgoing
3162 callout (which could be using a different interface). This has been
3163 changed to use the value of the helo_data option from the smtp transport
3164 instead - this is what is used when a message is actually being sent. If
3165 there is no remote transport (possible with a router that sets up host
3166 addresses), $smtp_active_hostname is used.
3168 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
3169 tweaks were necessary in order to get it to work (see also 21 below):
3170 (a) The code assumed that strncpy() returns a negative number on buffer
3171 overflow, which isn't the case. Replaced with Exim's string_format()
3173 (b) There were several signed/unsigned issues. I just did the minimum
3174 hacking in of casts. There is scope for a larger refactoring.
3175 (c) The code used strcasecmp() which is not a standard C function.
3176 Replaced with Exim's strcmpic() function.
3177 (d) The code set only $1; it now sets $auth1 as well.
3178 (e) A simple test gave the error "authentication client didn't specify
3179 service in request". It would seem that Dovecot has changed its
3180 interface. Fortunately there's a specification; I followed it and
3181 changed what the client sends and it appears to be working now.
3183 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
3186 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
3187 address A is aliased to B and C, where B exists and C does not. Without
3188 -v the output is "A verified" because verification stops after a
3189 successful redirection if more than one address is generated. However,
3190 with -v the child addresses are also verified. Exim was outputting "A
3191 failed to verify" and then showing the successful verification for C,
3192 with its parentage. It now outputs "B failed to verify", showing B's
3193 parentage before showing the successful verification of C.
3195 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
3196 look up a TXT record in a specific list after matching in a combined
3199 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
3200 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
3201 they consult the DNS. I had assumed they would set it the way they
3202 wanted; and indeed my experiments on Linux seem to show that in some
3203 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
3204 To be on the safe side, however, I have now made the interface to
3205 host_find_byname() similar to host_find_bydns(), with an argument
3206 containing the DNS resolver options. The host_find_byname() function now
3207 sets these options at its start, just as host_find_bydns() does. The smtp
3208 transport options dns_qualify_single and dns_search_parents are passed to
3209 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
3210 of host_find_byname() use the default settings of RES_DEFNAMES
3211 (qualify_single) but not RES_DNSRCH (search_parents).
3213 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
3214 spool_read_header() do less string testing, by means of a preliminary
3215 switch on the second character of optional "-foo" lines. (This is
3216 overdue, caused by the large number of possibilities that now exist.
3217 Originally there were few.) While I was there, I also converted the
3218 str(n)cmp tests so they don't re-test the leading "-" and the first
3219 character, in the hope this might squeeze out yet more improvement.
3221 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
3222 flag allowing group syntax was set by the header_syntax check but not
3223 turned off, possible causing trouble later; (2) The flag was not being
3224 set at all for the header_verify test, causing "group"-style headers to
3225 be rejected. I have now set it in this case, and also caused header_
3226 verify to ignore an empty address taken from a group. While doing this, I
3227 came across some other cases where the code for allowing group syntax
3228 while scanning a header line wasn't quite right (mostly, not resetting
3229 the flag correctly in the right place). These bugs could have caused
3230 trouble for malformed header lines. I hope it is now all correct.
3232 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
3233 with the "reply" argument non-NULL. The code, however (which originally
3234 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
3235 but it didn't always do it. This confused somebody who was copying the
3236 code for some other use. I have removed all the tests.
3238 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
3239 feature that was used to support insecure browsers during the U.S. crypto
3240 embargo. It requires special client support, and Exim is probably the
3241 only MTA that supported it -- and would never use it because real RSA is
3242 always available. This code has been removed, because it had the bad
3243 effect of slowing Exim down by computing (never used) parameters for the
3244 RSA_EXPORT functionality.
3246 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
3247 authenticator to fail if there's a tab character in the incoming data
3248 (there should never be unless someone is messing about, as it's supposed
3249 to be base64-encoded). Also added, on Timo's advice, the "secured" option
3250 if the connection is using TLS or if the remote IP is the same as the
3251 local IP, and the "valid-client-cert option" if a client certificate has
3254 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
3255 authenticators. This can be used for authorization after authentication
3256 succeeds. (In the case of plaintext, it servers for both authentication
3259 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
3260 if any retry times were supplied.
3262 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
3263 connection, where there is no client IP address to check. In this
3264 situation, the verify now always succeeds.
3266 PH/25 Applied John Jetmore's -Mset patch.
3268 PH/26 Added -bem to be like -Mset, but loading a message from a file.
3270 PH/27 In a string expansion for a processed (not raw) header when multiple
3271 headers of the same name were present, leading whitespace was being
3272 removed from all of them, but trailing whitespace was being removed only
3273 from the last one. Now trailing whitespace is removed from each header
3274 before concatenation. Completely empty headers in a concatenation (as
3275 before) are ignored.
3277 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
3278 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
3280 PH/29 [Removed. This was a change that I later backed out, and forgot to
3281 correct the ChangeLog entry (that I had efficiently created) before
3282 committing the later change.]
3284 PH/30 Exim was sometimes attempting to deliver messages that had suffered
3285 address errors (4xx response to RCPT) over the same connection as other
3286 messages routed to the same hosts. Such deliveries are always "forced",
3287 so retry times are not inspected. This resulted in far too many retries
3288 for the affected addresses. The effect occurred only when there were more
3289 hosts than the hosts_max_try setting in the smtp transport when it had
3290 the 4xx errors. Those hosts that it had tried were not added to the list
3291 of hosts for which the message was waiting, so if all were tried, there
3292 was no problem. Two fixes have been applied:
3294 (i) If there are any address or message errors in an SMTP delivery, none
3295 of the hosts (tried or untried) are now added to the list of hosts
3296 for which the message is waiting, so the message should not be a
3297 candidate for sending over the same connection that was used for a
3298 successful delivery of some other message. This seems entirely
3299 reasonable: after all the message is NOT "waiting for some host".
3300 This is so "obvious" that I'm not sure why it wasn't done
3301 previously. Hope I haven't missed anything, but it can't do any
3302 harm, as the worst effect is to miss an optimization.
3304 (ii) If, despite (i), such a delivery is accidentally attempted, the
3305 routing retry time is respected, so at least it doesn't keep
3306 hammering the server.
3308 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
3309 in ${readsocket because some servers need this prod.
3311 PH/32 Added some extra debug output when updating a wait-xxx database.
3313 PH/33 The hint "could be header name not terminated by colon", which has been
3314 given for certain expansion errors for a long time, was not being given
3315 for the ${if def:h_colon_omitted{... case.
3317 PH/34 The spec says: "With one important exception, whenever a domain list is
3318 being scanned, $domain contains the subject domain." There was at least
3319 one case where this was not true.
3321 PH/35 The error "getsockname() failed: connection reset by peer" was being
3322 written to the panic log as well as the main log, but it isn't really
3323 panic-worthy as it just means the connection died rather early on. I have
3324 removed the panic log writing for the ECONNRESET error when getsockname()
3327 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
3328 runs only) independently of the message's sender address. This meant
3329 that, if the 4xx error was in fact related to the sender, a different
3330 message to the same recipient with a different sender could confuse
3331 things. In particular, this can happen when sending to a greylisting
3332 server, but other circumstances could also provoke similar problems.
3333 I have changed the default so that the retry time for these errors is now
3334 based a combination of the sender and recipient addresses. This change
3335 can be overridden by setting address_retry_include_sender=false in the
3338 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
3339 remote server are returned as part of bounce messages. This was not
3340 happening for LMTP over a pipe (the lmtp transport), but now it is the
3341 same for both kinds of LMTP.
3343 PH/38 Despite being documented as not happening, Exim was rewriting addresses
3344 in header lines that were in fact CNAMEs. This is no longer the case.
3346 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
3347 and queue runs started by the daemon processed all messages. This has
3348 been fixed so that -R and -S can now usefully be given with -q<time>.
3350 PH/40 Import PCRE release 6.7 (fixes some bugs).
3352 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
3354 PH/42 Give an error if -q is specified more than once.
3356 PH/43 Renamed the variables $interface_address and $interface_port as
3357 $received_ip_address and $received_port, to make it clear that these
3358 values apply to message reception, and not to the outgoing interface when
3359 a message is delivered. (The old names remain recognized, of course.)
3361 PH/44 There was no timeout on the connect() call when using a Unix domain
3362 socket in the ${readsocket expansion. There now is.
3364 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
3365 be meaningful with "accept".
3367 SC/01 Eximstats V1.43
3368 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
3370 SC/02 Eximstats V1.44
3371 Use a glob alias rather than an array ref in the generated
3372 parser. This improves both readability and performance.
3374 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
3375 Collect SpamAssassin and rejection statistics.
3376 Don't display local sender or destination tables unless
3377 there is data to show.
3378 Added average volumes into the top table text output.
3380 SC/04 Eximstats V1.46
3381 Collect data on the number of addresses (recipients)
3382 as well as the number of messages.
3384 SC/05 Eximstats V1.47
3385 Added 'Message too big' to the list of mail rejection
3386 reasons (thanks to Marco Gaiarin).
3388 SC/06 Eximstats V1.48
3389 Mainlog lines which have GMT offsets and are too short to
3390 have a flag are now skipped.
3392 SC/07 Eximstats V1.49 (Alain Williams)
3393 Added the -emptyok flag.
3395 SC/08 Eximstats V1.50
3396 Fixes for obtaining the IP address from reject messages.
3398 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
3399 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
3400 whitespace changes from 4.64-PH/27
3402 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
3405 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
3406 are found, allow negative numbers in numeric criteria)
3408 JJ/06 exipick.20061117.2, added new $message_body_missing variable
3410 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
3411 to match changes made in 4.64-PH/43
3413 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
3415 PH/47 Put in an explicit test for a DNS lookup of an address record where the
3416 "domain" is actually an IP address, and force a failure. This locks out
3417 those revolvers/nameservers that support "A-for-A" lookups, in
3418 contravention of the specifications.
3420 PH/48 When a host name was looked up from an IP address, and the subsequent
3421 forward lookup of the name timed out, the host name was left in
3422 $sender_host_name, contrary to the specification.
3424 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
3425 restricted to single-key lookups, Exim was not diagnosing an error if
3426 * or *@ was used with a query-style lookup.
3428 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
3430 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
3431 long ago, but noone interested enough thought of it. Let's just say that
3432 the "1.1" means that there are some new functions that weren't there at
3433 some point in the past.
3435 PH/51 Error processing for expansion failure of helo_data from an smtp
3436 transport during callout processing was broken.
3438 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
3439 tested/used via the -bh/-bhc/-bs options.
3441 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
3442 bug, fixed in subsequent PCRE releases).
3444 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
3445 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
3447 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
3453 SC/01 Use a glob alias rather than an array ref in eximstats generated
3454 parser. This improves both readability and performance.
3456 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
3457 Don't display local sender or destination tables in eximstats unless
3458 there is data to show.
3459 Added average volumes into the eximstats top table text output.
3461 SC/03 Collect data on the number of addresses (recipients) as well
3462 as the number of messages in eximstats.
3464 TF/01 Correct an error in the documentation for the redirect router. Exim
3465 does (usually) call initgroups() when daemonizing.
3467 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
3468 with consistent privilege compared to when running as a daemon.
3470 TF/03 Note in the spec that $authenticated_id is not set for local
3471 submissions from trusted users.
3473 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
3474 Thanks to Dean Brooks <dean@iglou.com> for the patch.
3476 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
3477 by adding some example configuration directives to the default
3478 configuration file. A little bit of work is required to uncomment the
3479 directives and define how usernames and passwords are checked, but
3480 there is now a framework to start from.
3482 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
3483 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
3484 without this. I don't know how relevant this is to other LDAP libraries.
3486 PH/02 Add the verb name to the "unknown ACL verb" error.
3488 PH/03 Magnus Holmgren's patch for filter_prepend_home.
3490 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
3492 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
3493 directory not expanded when it should be if an expanded home directory
3494 was set for the address (which is overridden by the transport).
3496 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3499 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3500 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3501 because it is too late at that time, and has no effect.
3503 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3504 security issue with \' (bugzilla #107). I could not use the
3505 PQescapeStringConn() function, because it needs a PGconn value as one of
3508 PH/08 When testing addresses using -bt, indicate those final addresses that
3509 are duplicates that would not cause an additional delivery. At least one
3510 person was confused, thinking that -bt output corresponded to deliveries.
3511 (Suppressing duplicates isn't a good idea as you lose the information
3512 about possibly different redirections that led to the duplicates.)
3514 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3515 systems where poll() doesn't work, in particular OS X.
3517 PH/10 Added more information to debugging output for retry time not reached.
3519 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3520 operations in malware.c.
3522 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3525 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3526 syslog_duplication set false, log lines that would normally be written
3527 both the the main log and to the reject log were not written to syslog at
3530 PH/14 In the default configuration, change the use of "message" in ACL warn
3531 statements to "add_header".
3533 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3534 not followed by a command (e.g. "seen endif").
3536 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3537 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3540 PH/17 Added extra conditions to the default value of delay_warning_condition
3544 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3545 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3546 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3549 The Auto-Submitted: and various List- headers are standardised, whereas I
3550 don't think Precedence: ever was.
3552 PH/18 Refactored debugging code in route_finduser() to show more information,
3553 in particular, the error code if getpwnam() issues one.
3555 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3556 This is apparently needed in addition to the PH/07 change above to avoid
3557 any possible encoding problems.
3559 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3560 but not after initializing Perl.
3562 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3563 output them only if debugging. By default they are written stderr,
3564 apparently, which is not desirable.
3566 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3569 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3572 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3574 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3575 authenticated or an ident call has been made. Suppress the default
3576 values for $authenticated_id and $authenticated_sender (but permit -oMai
3577 and -oMas) when testing with -bh.
3579 PH/24 Re-jigged the order of the tests in the default configuration so that the
3580 tests for valid domains and recipients precede the DNS black list and CSA
3581 tests, on the grounds that those ones are more expensive.
3583 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3584 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3585 This bug exists in every version of Exim that I still have, right back to
3588 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3589 However, an attempt to turn on case-sensitivity in a regex key by
3590 including (?-i) didn't work because the subject string was already
3591 lowercased, and the effects were non-intuitive. It turns out that a
3592 one-line patch can be used to allow (?-i) to work as expected.
3598 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3599 other effects) broke the use of negated acl sub-conditions.
3601 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3604 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3605 "Deny" causes Exim to reject the incoming connection with a 554 error.
3606 Unfortunately, if there is a major crisis, such as a disk failure,
3607 tcp-wrappers gives "deny", whereas what one would like would be some
3608 kind of temporary error. A kludge has been added to help with this.
3609 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3610 554 error is used if errno is still zero or contains ENOENT (which occurs
3611 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3614 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3616 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3617 errors. Otherwise a message that provokes a temporary error (when other
3618 messages do not) can cause a whole host to time out.
3620 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3621 addresses were routed directly to files or pipes from a redirect router.
3622 File deliveries just didn't batch; pipe deliveries might have suffered
3625 PH/06 A failure to get a lock for a hints database would erroneously always say
3626 "Failed to get write lock", even when it was really a read lock.
3628 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3629 of 0600. This has been changed to use the value of the lockfile_mode
3630 option (which defaults to 0600).
3632 PH/08 Applied small patch from the Sieve maintainer.
3634 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3635 folder from quota calculations, a direct delivery into this folder messed
3636 up the contents of the maildirsize file. This was because the regex was
3637 used only to exclude .Trash (or whatever) when the size of the mailbox
3638 was calculated. There was no check that a delivery was happening into an
3639 excluded directory. This bug has been fixed by ignoring all quota
3640 processing for deliveries into excluded directories.
3642 PH/10 Added the maildirfolder_create_regex option to appendfile.
3648 PH/01 The code for finding all the local interface addresses on a FreeBSD
3649 system running IPv6 was broken. This may well have applied to all BSD
3650 systems, as well as to others that have similar system calls. The broken
3651 code found IPv4 interfaces correctly, but gave incorrect values for the
3652 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3653 that it would not match correctly against @[] and not recognize the IPv6
3656 PH/02 The ipliteral router was not recognizing addresses of the form user@
3657 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3659 PH/03 Added disable_ipv6.
3661 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3662 decoded form, because it is most often used to construct To: headers
3663 lines in autoreplies, and the decoded form may well be syntactically
3664 invalid. However, $reply_address has leading white space removed, and all
3665 newlines turned into spaces so that the autoreply transport does not
3668 PH/05 If group was specified without a user on a router, and no group or user
3669 was specified on a transport, the group from the router was ignored.
3671 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3672 for visible compile-time settings that can be used to change these
3673 numbers, for those that want even more. Backwards compatibility with old
3674 spool files has been maintained. However, going back to a previous Exim
3675 release will lost any variables that are in spool files.
3677 PH/07 Two small changes when running in the test harness: increase delay when
3678 passing a TCP/IP connection to a new process, in case the original
3679 process has to generate a bounce, and remove special handling of
3680 127.0.0.2 (sic), which is no longer necessary.
3682 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3683 be the same on different OS.
3685 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3688 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3689 whether --show-vars was specified or not
3691 JJ/02 exipick: Added support for new ACL variable spool format introduced
3694 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3695 syntactically invalid From: or Reply-to: line, and a filter used this to
3696 generate an autoreply, and therefore failed to obtain an address for the
3697 autoreply, Exim could try to deliver to a non-existent relative file
3698 name, causing unrelated and misleading errors. What now happens is that
3699 it logs this as a hard delivery error, but does not attempt to create a
3702 PH/11 The exinext utility has a -C option for testing purposes, but although
3703 the given file was scanned by exinext itself; it wasn't being passed on
3704 when Exim was called.
3706 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3707 an end-of-file indication when reading a command response.
3709 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3710 compiled. In many other places in Exim, IPv6 addresses are always
3711 recognized, so I have changed this. It also means that IPv4 domain
3712 literals of the form [IPV4:n.n.n.n] are now always recognized.
3714 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3715 used if the router is not running as root, for example, when verifying at
3716 ACL time, or when using -bh. The debugging output from this situation was
3717 non-existent - all you got was a failure to exec. I have made two
3720 (a) Failures to set uid/gid, the current directory, or a process leader
3721 in a subprocess such as that created by queryprogram now generate
3722 suitable debugging output when -d is set.
3724 (b) The queryprogram router detects when it is not running as root,
3725 outputs suitable debugging information if -d is set, and then runs
3726 the subprocess without attempting to change uid/gid.
3728 PH/15 Minor change to Makefile for building test_host (undocumented testing
3731 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3732 additional section of a DNS packet that returns MX or SRV records.
3733 Instead, it always explicitly searches for A/AAAA records. This avoids
3734 major problems that occur when a DNS server includes only records of one
3735 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3736 fixed another bug: if SRV records were looked up and the corresponding
3737 address records were *not* found in the additional section, the port
3738 values from the SRV records were lost.
3740 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3741 using the correct key (the original address) when searching the retry
3742 rules in order to find which one to use for generating the retry hint.
3744 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3745 adding the default one. Similarly, if it contains a Reply-To: header, the
3746 errors_reply_to option, if set, is not used.
3748 PH/19 When calculating a retry time, Exim used to measure the "time since
3749 failure" by looking at the "first failed" field in the retry record. Now
3750 it does not use this if it is later than than the arrival time of the
3751 message. Instead it uses the arrival time. This makes for better
3752 behaviour in cases where some deliveries succeed, thus re-setting the
3753 "first failed" field. An example is a quota failure for a huge message
3754 when small messages continue to be delivered. Without this change, the
3755 "time since failure" will always be short, possible causing more frequent
3756 delivery attempts for the huge message than are intended.
3757 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3759 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3760 $1, $2, $3) because the numerical variables can be reset during some
3761 expansion items (e.g. "match"), thereby losing the authentication data.
3763 PH/21 Make -bV show the size of off_t variables so that the test suite can
3764 decide whether to run tests for quotas > 2G.
3766 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3767 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3768 filecount value is greater than 2G or if a quota value is greater than 2G
3769 on a system where the size of off_t is not greater than 4, a panic error
3772 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3773 never match. The debug and -bh output now contains an explicit error
3774 message indicating a malformed IPv4 address or mask.
3776 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3777 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3778 PH/23 above applies.
3780 PH/25 Do not write to syslog when running in the test harness. The only
3781 occasion when this arises is a failure to open the main or panic logs
3782 (for which there is an explicit test).
3784 PH/26 Added the /no_tell option to "control=freeze".
3786 PH/27 If a host name lookup failed very early in a connection, for example, if
3787 the IP address matched host_lookup and the reverse lookup yielded a name
3788 that did not have a forward lookup, an error message of the form "no IP
3789 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3790 could be logged. Now it outputs the IP address instead of "NULL".
3792 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3793 allows the sender and the authenticated sender to be set when
3794 submitting a message from within Exim. Since child_open_exim() is
3795 documented for local_scan(), the new function should be too.
3797 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3798 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3799 results in an empty string is now treated as unset.
3801 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3803 PH/31 Added sender_verify_fail logging option.
3805 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3806 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3807 tidied the source and removed it altogether.
3809 PH/33 When a queue run was abandoned because the load average was too high, a
3810 log line was always written; now it is written only if the queue_run log
3811 selector is set. In addition, the log line for abandonment now contains
3812 information about the queue run such as the pid. This is always present
3813 in "start" and "stop" lines but was omitted from the "abandon" line.
3815 PH/34 Omit spaces between a header name and the colon in the error message that
3816 is given when verify = headers_syntax fails (if there are lots of them,
3817 the message gets confusing).
3819 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3820 names, as there are now some PTR records that contain slashes. This check
3821 is only to protect against broken name servers that fall over on strange
3822 characters, so the fact that it applies to all lookups doesn't matter.
3824 PH/36 Now that the new test suite is complete, we can remove some of the
3825 special code in Exim that was needed for the old test suite. For example,
3826 sorting DNS records because real resolvers return them in an arbitrary
3827 order. The new test suite's fake resolver always returns records in the
3830 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3831 bounces) except when queue_only is set, to avoid logging races between
3832 the different processes.
3834 PH/38 Panic-die if .include specifies a non-absolute path.
3836 PH/39 A tweak to the "H" retry rule from its user.
3838 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3839 a label. They prevented compilation on older perls.
3841 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3842 a warning to be raised on newish perls.
3844 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3845 on queue. Changes to match documented behaviour of showing count of
3846 messages matching specified criteria.
3848 PH/40 Changed the default ident timeout from 30s to 5s.
3850 PH/41 Added support for the use of login_cap features, on those BSD systems
3851 that have them, for controlling the resources used by pipe deliveries.
3853 PH/42 The content-scanning code uses fopen() to create files in which to put
3854 message data. Previously it was not paying any attention to the mode of
3855 the files. Exim runs with umask(0) because the rest of the code creates
3856 files with open(), and sets the required mode explicitly. Thus, these
3857 files were ending up world-writeable. This was not a big issue, because,
3858 being within the spool directory, they were not world-accessible. I have
3859 created a function called modefopen, which takes an additional mode
3860 argument. It sets umask(777), creates the file, chmods it to the required
3861 mode, then resets the umask. All the relevant calls to fopen() in the
3862 content scanning code have been changed to use this function.
3864 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3865 to 24 hours. This avoids potential overflow problems when processing G
3866 and H retry rules. I suspect nobody ever tinkers with this value.
3868 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3870 PH/45 When the plaintext authenticator is running as a client, the server's
3871 challenges are checked to ensure they are valid base64 strings. By
3872 default, the authentication attempt is cancelled if an invalid string is
3873 received. Setting client_ignore_invalid_base64 true ignores these errors.
3874 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3875 they are received. Thus, the responses can be made to depend on the
3876 challenges. If an invalid string is ignored, an empty string is placed in
3879 PH/46 Messages that are created by the autoreply transport now contains a
3880 References: header, in accordance with RFCs 2822 and 3834.
3882 PH/47 Added authenticated_sender_force to the smtp transport.
3884 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3886 PH/49 Installed latest patch from the Sieve maintainer.
3888 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3889 was also set, the appendfile transport was unnecessarily scanning a
3890 directory of message files (e.g. for maildir delivery) to find the count
3891 of files (along with the size), even though it did not need this
3892 information. It now does the scan only if it needs to find either the
3893 size of the count of files.
3895 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3897 PH/52 Two bugs concerned with error handling when the smtp transport is
3900 (i) Exim was not creating retry information for temporary errors given
3901 for individual recipients after the DATA command when the smtp transport
3902 was used in LMTP mode. This meant that they could be retried too
3903 frequently, and not timed out correctly.
3905 (ii) Exim was setting the flag that allows error details to be returned
3906 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3907 recipients that were returned after the DATA command.
3909 PH/53 This is related to PH/52, but is more general: for any failing address,
3910 when detailed error information was permitted to be returned to the
3911 sender, but the error was temporary, then after the final timeout, only
3912 "retry timeout exceeded" was returned. Now it returns the full error as
3913 well as "retry timeout exceeded".
3915 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3916 do this, and (what is worse) MTAs that accept it.
3918 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3919 will now be deprecated.
3921 PH/56 New os.c-cygwin from the Cygwin maintainer.
3923 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3924 formats (previously only available in exim formats via -bpr, -bpru,
3925 and -bpra. Now also available in native and exiqgrep formats)
3927 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3928 with very large, slow to parse queues
3930 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3932 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3934 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3935 responses to authentication challenges, though it was showing the
3936 challenges; (ii) I've removed the CR characters from the debug output for
3939 PH/58 Allow for the insertion of a newline as well as a space when a string
3940 is turned into more than one encoded-word during RFC 2047 encoding. The
3941 Sieve code now uses this.
3943 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3944 data_4xx, lost_connection, tls_required.
3946 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3947 message was being sent as an SMTP response.
3949 PH/61 Add -l and -k options to exicyclog.
3951 PH/62 When verifying, if an address was redirected to one new address, so that
3952 verification continued, and the new address failed or deferred after
3953 having set something in $address_data, the value of $address_data was not
3954 passed back to the ACL. This was different to the case when no
3955 redirection occurred. The value is now passed back in both cases.
3957 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3958 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3959 use login_cap.h, so on its own it isn't the distinguishing feature. The
3960 new name refers directly to the setclassresources() function.
3962 PH/65 Added configuration files for NetBSD3.
3964 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3966 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3967 is preferred over IPv4.
3969 PH/68 The bounce_return_message and bounce_return_body options were not being
3970 honoured for bounces generated during the reception of non-SMTP messages.
3971 In particular, this applied to messages rejected by the ACL. This bug has
3972 been fixed. However, if bounce_return_message is true and bounce_return_
3973 body is false, the headers that are returned for a non-SMTP message
3974 include only those that have been read before the error was detected.
3975 (In the case of an ACL rejection, they have all been read.)
3977 PH/69 The HTML version of the specification is now built in a directory called
3978 spec_html instead of spec.html, because the latter looks like a path with
3979 a MIME-type, and this confuses some software.
3981 PH/70 Catch two compiler warnings in sieve.c.
3983 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3984 function verify_get_ident() calls ip_connect() to connect a socket, but
3985 if the "connect()" function timed out, ip_connect() used to close the
3986 socket. However, verify_get_ident() also closes the socket later, and in
3987 between Exim writes to the log, which may get opened at this point. When
3988 the socket was closed in ip_connect(), the log could get the same file
3989 descriptor number as the socket. This naturally causes chaos. The fix is
3990 not to close the socket in ip_connect(); the socket should be closed by
3991 the function that creates it. There was only one place in the code where
3992 this was missing, in the iplookup router, which I don't think anybody now
3993 uses, but I've fixed it anyway.
3995 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3996 well as to direct DNS lookups. Otherwise the handling of names in host
3997 lists is inconsistent and therefore confusing.
4003 PH/01 Two changes to the default runtime configuration:
4005 (1) Move the checks for relay_from_hosts and authenticated clients from
4006 after to before the (commented out) DNS black list checks.
4008 (2) Add control=submission to the relay_from_hosts and authenticated
4009 clients checks, on the grounds that messages accepted by these
4010 statements are most likely to be submissions.
4012 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
4014 (1) Generate an error if the third argument for the ${prvs expansion is
4017 (2) Treat a missing third argument of ${prvscheck as if it were an empty
4020 (3) Reset the variables that are obtained from the first argument of
4021 ${prvscheck and used in the second argument before leaving the code,
4022 because their memory is reclaimed, so using them afterwards may do
4025 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
4026 one (it's much easier than Tom thought :-).
4028 (5) Because of (4), we can now allow for the use of $prvscheck_result
4029 inside the third argument.
4031 PH/03 For some reason, the default setting of PATH when running a command from
4032 a pipe transport was just "/usr/bin". I have changed it to
4035 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
4036 anything to be listed in the output from -bV.
4038 PH/05 When a filter generated an autoreply, the entire To: header line was
4039 quoted in the delivery log line, like this:
4041 => >A.N.Other <ano@some.domain> <original@ddress> ...
4043 This has been changed so that it extracts the operative address. There
4044 may be more than one such address. If so, they are comma-separated, like
4047 => >ano@some.domain,ona@other.domain <original@ddress> ...
4049 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
4050 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
4051 not being looked up in the rDNS to get a host name, Exim was showing the
4052 IP address twice in Received: lines, even though the IP addresses were
4053 identical. For example:
4055 Received: from [1.2.3.4] (helo=[1.2.3.4])
4057 However, if the real host name was known, it was omitting the HELO data
4058 if it matched the actual IP address. This has been tidied up so that it
4059 doesn't show the same IP address twice.
4061 PH/07 When both +timestamp and +memory debugging was on, the value given by
4062 $tod_xxx expansions could be wrong, because the tod_stamp() function was
4063 called by the debug printing, thereby overwriting the timestamp buffer.
4064 Debugging no longer uses the tod_stamp() function when +timestamp is set.
4066 PH/08 When the original message was included in an autoreply transport, it
4067 always said "this is a copy of the message, including all the headers",
4068 even if body_only or headers_only was set. It now gives an appropriate
4071 PH/09 Applied a patch from the Sieve maintainer which:
4073 o fixes some comments
4074 o adds the (disabled) notify extension core
4075 o adds some debug output for the result of if/elsif tests
4076 o points to the current vacation draft in the documentation
4077 and documents the missing references header update
4081 o fixes a bug in processing the envelope test (when testing
4082 multiple envelope elements, the last element determined the
4085 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
4086 Electronic Mail") by including:
4088 Auto-submitted: auto-generated
4090 in the messages that it generates (bounce messages and others, such as
4091 warnings). In the case of bounce messages for non-SMTP messages, there was
4092 also a typo: it was using "Auto_submitted" (underscore instead of
4093 hyphen). Since every message generated by Exim is necessarily in response
4094 to another message, thes have all been changed to:
4096 Auto-Submitted: auto-replied
4098 in accordance with these statements in the RFC:
4100 The auto-replied keyword:
4102 - SHOULD be used on messages sent in direct response to another
4103 message by an automatic process,
4105 - MUST NOT be used on manually-generated messages,
4107 - MAY be used on Delivery Status Notifications (DSNs) and Message
4108 Disposition Notifications (MDNs),
4110 - MUST NOT be used on messages generated by automatic or periodic
4111 processes, except for messages which are automatic responses to
4114 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
4115 to the default Received: header definition.
4117 PH/12 Added log selector acl_warn_skipped (default on).
4119 PH/13 After a successful wildlsearch lookup, discard the values of numeric
4120 variables because (a) they are in the wrong storage pool and (b) even if
4121 they were copied, it wouldn't work properly because of the caching.
4123 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
4124 checking when decoding. Apparently there are clients that generate
4125 overlong encoded strings. Why am I not surprised?
4127 PH/15 If the first argument of "${if match_address" was not empty, but did not
4128 contain an "@" character, Exim crashed. Now it writes a panic log message
4129 and treats the condition as false.
4131 PH/16 In autoreply, treat an empty string for "once" the same as unset.
4133 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
4134 extension "envelope-auth". The code is finished and in agreement with
4135 other implementations, but there is no documentation so far and in fact,
4136 nobody wrote the draft yet. This extension is currently #undef'ed, thus
4137 not changing the active code.
4139 Print executed "if" and "elsif" statements when debugging is used. This
4140 helps a great deal to understand what a filter does.
4142 Document more things not specified clearly in RFC3028. I had all this
4143 sorted out, when out of a sudden new issues came to my mind. Oops."
4145 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
4148 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
4149 canonical form (as documented). However, after a host name lookup from
4150 the IP address, check_host() was doing a simple string comparison with
4151 addresses acquired from the DNS when checking that the found name did
4152 have the original IP as one of its addresses. Since any found IPv6
4153 addresses are likely to be in abbreviated form, the comparison could
4154 fail. Luckily, there already exists a function for doing the comparison
4155 by converting both addresses to binary, so now that is used instead of
4156 the text comparison.
4158 PH/20 There was another similar case to PH/19, when a complete host name was
4159 given in a host list; looking up its IP address could give an abbreviated
4160 form, whereas the current host's name might or might not be abbreviated.
4161 The same fix has been applied.
4167 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
4168 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
4171 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
4172 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
4174 PH/03 Typo: missing ".o" in src/pcre/Makefile.
4176 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
4177 header line, restrict the check to what is listed in RFCs 2369 and 2929.
4178 Also, for "Auto-Submitted", treat anything other than "no" as
4179 non-personal, in accordance with RFC 3834. (Previously it treated
4180 anything starting "auto-" as non-personal.)
4182 TF/01 The control=submission/name=... option had a problem with syntax
4183 errors if the name included a slash character. The /name= option
4184 now slurps the rest of the string, so it can include any characters
4185 but it must come last in the list of options (after /sender_retain
4188 PH/05 Some modifications to the interface to the fake nameserver for the new
4196 TK/01 Added the "success_on_redirect" address verification option. See
4197 NewStuff for rationale and an example.
4199 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
4201 PH/02 Patch to exigrep to allow it to work on syslog lines.
4203 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
4204 fread() to skip over the body file's header line, because in Cygwin the
4205 header line is locked and is inaccessible.
4207 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
4208 co-exist for some time) to make it clear that it is the Exim ID that is
4209 referenced, not the Message-ID: header line.
4211 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
4212 string_format() function, because snprintf() does not exist on all
4215 PH/06 The use of forbid_filter_existstest now also locks out the use of the
4216 ${stat: expansion item.
4218 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
4219 protocol synchronization error", to keep the pedants happy.
4221 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
4222 well as for IRIX systems, when gcc is being used. See the host.c source
4225 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
4227 PH/10 Named domain lists were not working if used in a queue_smtp_domains
4230 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
4231 transport and to the smtp transport in LMTP mode.
4233 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
4235 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
4236 run a filter in a subprocess. This could lead to confusion in subsequent
4237 lookups in the parent process. There should also be a search_tidyup() at
4238 the end of the subprocess.
4240 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
4241 only if the host matched helo_try_verify_hosts, which caused the
4242 verification to occur when the EHLO/HELO command was issued. The ACL just
4243 tested the remembered result. Now, if a previous verification attempt has
4244 not happened, "verify = helo" does it there and then.
4246 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
4248 TK/03 Fix log output including CR from clamd.
4250 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
4251 exist provoked a memory error which could cause a segfault.
4253 PH/15 Installed PCRE 6.2
4255 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
4257 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
4258 of the problem. Specifically, suggested +O2 rather than +O1 for the
4261 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
4263 PH/20 If a delivery was routed to a non-standard port by means of an SRV
4264 record, the port was not correctly logged when the outgoing_port log
4265 selector was set (it logged the transort's default port).
4267 PH/21 Added support for host-specific ports to manualroute, queryprogram,
4268 fallback_hosts, and "hosts" in the smtp transport.
4270 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
4271 host errors such as "Connection refused".
4273 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
4274 authentication with radiusclient 0.4.9:
4276 - Error returned from rc_read_config was caught wrongly
4277 - Username/password not passed on to radius server due to wrong length.
4279 The presumption is that some radiusclient API changes for 4.51/PH/17
4280 were not taken care of correctly. The code is still untested by me (my
4281 Linux distribution still has 0.3.2 of radiusclient), but it was
4282 contributed by a Radius user.
4284 PH/24 When doing a callout, the value of $domain wasn't set correctly when
4285 expanding the "port" option of the smtp transport.
4287 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
4288 while reading a MIME header. Thanks to Tom Hughes for a patch.
4290 PH/24 Include config.h inside local_scan.h so that configuration settings are
4293 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
4294 that in an ACL for RCPT (for example), you can examine exactly what was
4297 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
4298 commands, but it was not correctly comparing the address with the actual
4299 client host address. Thus, it would show the EHLO address in Received:
4300 header lines when this was not necessary.
4302 PH/27 Added the % operator to ${eval:}.
4304 PH/28 Exim tries to create and chdir to its spool directory when it starts;
4305 it should be ignoring failures (because with -C, for example, it has lost
4306 privilege). It wasn't ignoring creation failures other than "already
4309 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
4312 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
4313 because an input line was too long, either on its own, or by virtue of
4314 too many continuations, the temporary file was not being removed, and the
4315 return code was incorrect.
4317 PH/31 Missing "BOOL" in function definition in filtertest.c.
4319 PH/32 Applied Sieve patches from the maintainer.
4321 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
4323 PH/33 Added "verify = not_blind".
4325 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
4326 Local/Makefile (with some defaults set). These are used in built scripts
4327 such as exicyclog, but they have never been used in the exim_install
4328 script (though there are many overriding facilities there). I have
4329 arranged that the exim_install script now takes note of these two
4332 PH/35 Installed configuration files for Dragonfly.
4334 PH/36 When a locally submitted message by a trusted user did not contain a
4335 From: header, and the sender address was obtained from -f or from an SMTP
4336 MAIL command, and the trusted user did not use -F to supply a sender
4337 name, $originator_name was incorrectly used when constructing a From:
4338 header. Furthermore, $originator_name was used for submission mode
4339 messages from external hosts without From: headers in a similar way,
4340 which is clearly wrong.
4342 PH/37 Added control=suppress_local_fixups.
4344 PH/38 When log_selector = +received_sender was set, and the addition of the
4345 sender made the log line's construction buffer exactly full, or one byte
4346 less than full, an overflow happened when the terminating "\n" was
4349 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
4350 when the result of a list match is failure because a DNS lookup failed.
4352 PH/40 RM_COMMAND is now used in the building process.
4354 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
4355 the "build-* directories that it finds.
4357 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
4358 address in a domain literal was a prefix of an interface address.
4360 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
4361 when verifying a sender address, unless rewrite_headers is false.
4363 PH/44 Wrote a long comment about why errors_to addresses are verified as
4364 recipients, not senders.
4366 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
4367 the ratelimit ACL was added.
4369 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
4371 PH/46 Added extra information about PostgreSQL errors to the error string.
4373 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
4374 avoiding the need to install special zones in a real server. This is
4375 backwards compatible; if it can't find the fake resolver, it drops back.
4376 Thus, both old and new test suites can be run.
4378 TF/02 Added util/ratelimit.pl
4380 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
4381 clock is set back in time.
4383 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
4384 Candler <B.Candler@pobox.com>.
4386 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
4387 OK for addresses that are the result of redirections.
4389 PH/48 A number of further additions for the benefit of the new test suite,
4390 including a fake gethostbyname() that interfaces to the fake DNS resolver
4393 TF/06 The fix for widen_domains has also been applied to qualify_single and
4394 search_parents which are the other dnslookup options that can cause
4397 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
4400 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
4402 TF/07 Exim produced the error message "an SRV record indicated no SMTP
4403 service" if it encountered an MX record with an empty target hostname.
4404 The message is now "an MX or SRV record indicated no SMTP service".
4406 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
4407 if the DNS of the sending site is misconfigured. This is quite a
4408 common situation. This change restores the behaviour of treating a
4409 helo verification defer as a failure.
4411 PH/51 If self=fail was set on a router, the bounce message did not include the
4412 actual error message.
4418 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
4420 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
4421 command itself ended in error, the underlying message about the transport
4422 filter timeout was being overwritten with the pipe command error. Now the
4423 underlying error message should be appended to the second error message.
4425 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
4427 PH/02 Reduce the amount of output that "make" produces by default. Full output
4428 can still be requested.
4430 PH/03 The warning log line about a condition test deferring for a "warn" verb
4431 was being output only once per connection, rather than after each
4432 occurrence (because it was using the same function as for successful
4433 "warn" verbs). This seems wrong, so I have changed it.
4435 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
4436 it should not have, which might have caused a crash in the right
4437 circumstances, but probably never did.
4439 PH/04 Installed a modified version of Tony Finch's patch to make submission
4440 mode fix the return path as well as the Sender: header line, and to
4441 add a /name= option so that you can make the user's friendly name appear
4444 TF/03 Added the control = fakedefer ACL modifier.
4446 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
4447 Mark Lowes for thorough testing.
4449 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
4451 TK/03 Merged latest SRS patch from Miles Wilton.
4453 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
4454 with the definition in sysexits.h (which is #included earlier).
4455 Fortunately, Exim does not actually use EX_OK. The code used to try to
4456 preserve the sysexits.h value, by assuming that macro definitions were
4457 scanned for macro replacements. I have been disabused of this notion,
4458 so now the code just undefines EX_OK before #including unistd.h.
4460 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
4461 in the smtp transport. When a block could not be written in a single
4462 write() function, the timeout was being re-applied to each part-write.
4463 This seems wrong - if the receiver was accepting one byte at a time it
4464 would take for ever. The timeout is now adjusted when this happens. It
4465 doesn't have to be particularly precise.
4467 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
4468 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
4470 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
4471 without a domain if the check to <postmaster@domain> fails.
4473 SC/01 Eximstats: added -xls and the ability to specify output files
4474 (patch written by Frank Heydlauf).
4476 SC/02 Eximstats: use FileHandles for outputting results.
4478 SC/03 Eximstats: allow any combination of xls, txt, and html output.
4480 SC/04 Eximstats: fixed display of large numbers with -nvr option
4482 SC/05 Eximstats: fixed merging of reports with empty tables.
4484 SC/06 Eximstats: added the -include_original_destination flag
4486 SC/07 Eximstats: removed tabs and trailing whitespace.
4488 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
4490 TK/06 MBOX spool code: Add real "From " MBOX separator line
4491 so the .eml file is really in mbox format (even though
4492 most programs do not really care). Patch from Alex Miller.
4494 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4495 The latter is generated from $received_to and is only set if the
4496 message has one envelope recipient. SA can use these headers,
4497 obviously out-of-the-box. Patch from Alex Miller.
4499 PH/08 The ${def test on a variable was returning false if the variable's
4500 value was "0", contrary to what the specification has always said!
4501 The result should be true unless the variable is empty.
4503 PH/09 The syntax error of a character other than { following "${if
4504 def:variable_name" (after optional whitespace) was not being diagnosed.
4505 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4506 accidental colon was present, for example, could give incorrect results.
4508 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4509 result is used (not including appendfile, where other changes are about
4512 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4513 This involved changing a lot of size variables from int to off_t. It
4514 should work with maildirs and everything.
4516 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4517 spamd dying while we are connected to it.
4519 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4522 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4523 function for BDB 4.3.
4525 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4527 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4528 the sources for PCRE have been re-arranged and more files are now
4531 PH/15 The code I had for printing potentially long long variables in PH/11
4532 above was not the best (it lost precision). The length of off_t variables
4533 is now inspected at build time, and an appropriate printing format (%ld
4534 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4535 to be "long long int" or "long int". This is needed for the internal
4536 formatting function string_vformat().
4538 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4539 the configuration file to be ":syslog", then the script "guesses" where
4540 the logs files are, rather than using the compiled in default. In our
4541 case the guess is not the same as the compiled default, so the script
4542 suddenly stopped working when I started to use syslog. The patch checks
4543 to see if log_file_path is "". If so, it attempts to read it from exim
4544 with no configuration file to get the compiled in version, before it
4545 falls back to the previous guessing code."
4547 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4548 implementing BATV in an Exim configuration. See NewStuff for the gory
4551 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4552 Makefile that are specific to HP-UX.
4554 PH/18 If the "use_postmaster" option was set for a recipient callout together
4555 with the "random" option, the postmaster address was used as the MAIL
4556 FROM address for the random test, but not for the subsequent recipient
4557 test. It is now used for both.
4559 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4560 patch removes a few documentation additions to RFC 3028, because the
4561 latest draft now contains them. It adds the new en;ascii-case comparator
4562 and a new error check for 8bit text in MIME parts. Comparator and
4563 require names are now matched exactly. I enabled the subaddress
4564 extension, but it is not well tested yet (read: it works for me)."
4566 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4567 rework some of the code of TK/09 above to avoid the hardwired use of
4568 "%lld" and "long long". Replaced the call to snprintf() with a call to
4571 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4572 records point to non-existent hosts", "retry timeout exceeded", and
4573 "retry time not reached for any host after a long failure period".
4575 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4576 experimental DomainKeys support:
4578 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4579 (2) On an error such as an illegally used "control", the wrong name for
4580 the control was given.
4582 These problems did NOT occur unless DomainKeys support was compiled.
4584 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4586 PH/24 Added ${if match_ip condition.
4588 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4589 databases so that it will be absolutely obvious if a crash occurs in the
4590 DB library. This is a regular occurrence (often caused by mis-matched
4593 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4594 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4595 picked up on a user's system that detects such things. There doesn't seem
4596 to be a gcc warning option for this - only an attribute that has to be
4597 put on the function's prototype. It seems that in Fedora Core 4 they have
4598 set this on a number of new functions. No doubt there will be more in due
4601 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4602 specify a transport. However, if an address that was verified by such a
4603 router was the subject of a callout, Exim crashed because it tried to
4604 read the rcpt_include_affixes from the non-existent transport. Now it
4605 just assumes that the setting of that option is false. This bug was
4606 introduced by 4.51/PH/31.
4608 PH/28 Changed -d+all to exclude +memory, because that information is very
4609 rarely of interest, but it makes the output a lot bigger. People tend to
4610 do -d+all out of habit.
4612 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4613 code in os-type was giving problems when libc.so lives in lib64, like on
4616 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4617 aren't the modern standard, and it seems that some systems' include files
4618 don't always have them. Exim was already checking for some of the newer
4619 ones like T_AAAA, and defining it itself. I've added checks for all the
4620 record types that Exim uses.
4622 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4623 not automatically generating a new one, as it is supposed to. This
4624 prevented TLS from working. If the file did exist, but contained invalid
4625 data, a new version was generated, as expected. It was only the case of a
4626 non-existent file that was broken.
4628 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4629 with a change in libdomainkeys > 0.64.
4631 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4632 from DNS. If the selector record carries the flag, it now has
4633 precedence over the domain-wide flag.
4635 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4637 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4638 the use of a port name that isn't defined in /etc/services) occurred, the
4639 message was deferred as in a normal delivery, and thus remained on the
4640 spool, instead of being failed because of the mua_wrapper setting. This
4641 is now fixed, and I tidied up some of the mua_wrapper messages at the
4644 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4645 the messages in a hash of arrays rather than using individual hashes.
4646 This is a bit cleaner and results in dramatic memory savings, albeit
4647 at a slight CPU cost.
4649 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4650 as requested by Marc Sherman.
4652 SC/10 Eximstats: added histograms for user specified patterns as requested
4655 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4657 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4658 fopen() in the content-scanning modules that did not already have it.
4664 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4665 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4667 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4669 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4671 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4672 to test Sieve filters that use "vacation".
4674 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4675 that changes the way the GnuTLS parameters are stored in the cache file.
4676 The new format can be generated externally. For backward compatibility,
4677 if the data in the cache doesn't make sense, Exim assumes it has read an
4678 old-format file, and it generates new data and writes a new file. This
4679 means that you can't go back to an older release without removing the
4682 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4683 work if there are any delivery delays because "one_time" forces the
4684 parent to be marked "delivered", so its unseen clone is never tried
4685 again. For this reason, Exim now forbids the simultaneous setting of
4688 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4689 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4690 entry if you want to know the details. The fix, however, neglected to
4691 consider the case when local delivery batching is involved. The test for
4692 "previously delivered" was not happening when checking to see if an
4693 address could be batched with a previous (undelivered) one; under
4694 certain circumstances this could lead to multiple deliveries to the same
4697 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4698 in its include files, and this causes problems building Exim.
4700 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4701 header_syntax) but Exim was just ignoring anything given after a slash.
4702 In particular, this caused confusion with an attempt to use "verify =
4703 reverse_host_lookup/defer_ok". An error is now given when options are
4704 supplied for verify items that do not have them. (Maybe reverse_host_
4705 lookup should have a defer_ok option, but that's a different point.)
4707 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4708 defined by RFC 821) to 2048, because there were problems with some AUTH
4709 commands, and RFC 1869 says the size should be increased for extended
4710 SMTP commands that take arguments.
4712 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4715 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4716 "unknown" error; now it says that the functionality isn't in the binary.
4718 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4719 an address' error message when a string expansion fails (syntax or
4720 whatever). Otherwise the password may appear in the log. Following change
4721 PH/42 below, there is no longer a chance of it appearing in a bounce
4724 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4726 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4727 of its addresses was ever tried. (Bugzilla bug #2.)
4729 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4730 the result incorrectly in the debug output. (It correctly added a newline
4731 to what was transported.)
4733 TF/01 Added $received_time.
4735 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4736 commented out examples of how to interface to a virus scanner and to
4737 SpamAssassin. Also added commented examples of av_scanner and
4738 spamd_address settings.
4740 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4741 and controls are allowed in which ACLs. There were a couple of minor
4742 errors. Some of the entries in the conditions table (which is a table of
4743 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4744 negation of where the condition IS allowed.
4746 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4748 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4749 header file does not have a version number, so I've had to invent a new
4750 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4751 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4752 radiusclient), but it was contributed by a Radius user.
4754 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4755 files or directories, for OpenSSL.
4757 PH/19 When an Exim process that is running as root has to create an Exim log
4758 file, it does so in a subprocess that runs as exim:exim so as to get the
4759 ownership right at creation (otherwise, other Exim processes might see
4760 the file with the wrong ownership). There was no test for failure of this
4761 fork() call, which would lead to the process getting stuck as it waited
4762 for a non-existent subprocess. Forks do occasionally fail when resources
4763 run out. I reviewed all the other calls to fork(); they all seem to check
4766 PH/20 When checking for unexpected SMTP input at connect time (before writing
4767 the banner), Exim was not dealing correctly with a non-positive return
4768 from the read() function. If the client had disconnected by this time,
4769 the result was a log entry for a synchronization error with an empty
4770 string after "input=" when read() returned zero. If read() returned -1
4771 (an event I could not check), uninitialized data bytes were printed.
4772 There were reports of junk text (parts of files, etc) appearing after
4775 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4777 PH/22 Added support for macro redefinition, and (re)definition in between
4778 driver and ACL definitions.
4780 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4781 forgetting to use the resulting value; it was using the unexpanded value.
4783 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4784 hadn't been configured. The fix is from Juergen Kreileder, who
4785 understands it better than I do:
4787 "Here's what I see happening with three configured cyrus_sasl
4788 authenticators configured (plain, login, cram-md5):
4790 On startup auth_cyrus_sasl_init() gets called for each of these.
4791 This means three calls to sasl_listmech() without a specified mech_list.
4792 => SASL tests which mechs of all available mechs actually work
4793 => three warnings about OTP not working
4794 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4796 With the patch, sasl_listmech() also gets called three times. But now
4797 SASL's mech_list option is set to the server_mech specified in the the
4798 authenticator. Or in other words, the answer from sasl_listmech()
4799 gets limited to just the mech you're testing for (which is different
4801 => the return list contains just 'plain' or 'login', 'cram-md5' or
4802 nothing depending on the value of ob->server_mech.
4804 I've just tested the patch: Authentication still works fine,
4805 unavailable mechs specified in the exim configuration are still
4806 caught, and the auth.log warnings about OTP are gone."
4808 PH/25 When debugging is enabled, the contents of the command line are added
4809 to the debugging output, even when log_selector=+arguments is not
4812 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4813 answer is "GNU", and only if the return is "GNU/something" is the answer
4816 PH/27 $acl_verify_message is now set immediately after the failure of a
4817 verification in an ACL, and so is available in subsequent modifiers. In
4818 particular, the message can be preserved by coding like this:
4820 warn !verify = sender
4821 set acl_m0 = $acl_verify_message
4823 Previously, $acl_verify_message was set only while expanding "message"
4824 and "log_message" when a very denied access.
4826 PH/28 Modified OS/os.c-Linux with
4828 -#ifndef OS_LOAD_AVERAGE
4829 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4831 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4832 nomenclature these days.)
4834 PH/29 Installed patch from the Sieve maintainer that adds the options
4835 sieve_useraddress and sieve_subaddress to the redirect router.
4837 PH/30 In these circumstances:
4838 . Two addresses routed to the same list of hosts;
4839 . First host does not offer TLS;
4840 . First host accepts first address;
4841 . First host gives temporary error to second address;
4842 . Second host offers TLS and a TLS session is established;
4843 . Second host accepts second address.
4844 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4845 and peerdn, if requested) that were in fact used only for the second
4848 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4849 attention to any local part prefix or suffix that was matched by the
4850 router that accepted the address. It now behaves in the same way as it
4851 does for delivery: the affixes are removed from the local part unless
4852 rcpt_include_affixes is set on the transport.
4854 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4855 timeout during the DATA phase of an incoming message.
4857 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4858 applied a patch sanctioned by the Sieve maintainer.
4860 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4861 the uid or gid is negative. A case of a negative gid caused this to be
4862 noticed. The fix allows for either to be negative.
4864 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4865 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4868 PH/36 The change PH/12 above was broken. Fixed it.
4870 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4871 the grounds that routing the same address twice would always produce the
4872 same answer. This might have been true once, but it is certainly no
4873 longer true now. Routing a child address may depend on the previous
4874 routing that produced that child. Some complicated redirection strategies
4875 went wrong when messages had multiple recipients, and made Exim's
4876 behaviour dependent on the order in which the addresses were given.
4878 I have moved the duplicate checking until after the routing is complete.
4879 Exim scans the addresses that are assigned to local and remote
4880 transports, and removes any duplicates. This means that more work will be
4881 done, as duplicates will always all be routed, but duplicates are
4882 presumably rare, so I don't expect this is of any significance.
4884 For deliveries to pipes, files, and autoreplies, the duplicate checking
4885 still happens during the routing process, since they are not going to be
4888 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4889 It corrects a timeout issue with spamd. This is Ian's comment: "The
4890 background is that sometimes spamd either never reads data from a
4891 connection it has accepted, or it never writes response data. The exiscan
4892 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4893 blindly assumes that writes won't block so it may never time out."
4895 PH/39 Allow G after quota size as well as K and M.
4897 PH/40 The value set for $authenticated_id in an authenticator may not contain
4898 binary zeroes or newlines because the value is written to log lines and
4899 to spool files. There was no check on this. Now the value is run through
4900 the string_printing() function so that such characters are converted to
4901 printable escape sequences.
4903 PH/41 $message_linecount is a new variable that contains the total number of
4904 lines in the message. Compare $body_linecount, which is the count for the
4907 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4908 bounce and delay warning messages, except in certain special cases, which
4911 (a) An SMTP error message from a remote host;
4912 (b) A message specified in a :fail: redirection;
4913 (c) A message specified in a "fail" command in a system filter;
4914 (d) A message specified in a FAIL return from the queryprogram router;
4915 (e) A message specified by the cannot_route_message router option.
4917 In these cases only, Exim does include the error details in bounce and
4918 warning messages. There are also a few cases where bland messages such
4919 as "unrouteable address" or "local delivery error" are given.
4921 PH/43 $value is now also set for the "else" part of a ${run expansion.
4923 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4924 being worked on, but at least Exim now implements the latest version to
4927 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4928 process to complete was treated as a delivery failure, a timeout while
4929 writing the message to the pipe was logged, but erroneously treated as a
4930 successful delivery. Such timeouts include transport filter timeouts. For
4931 consistency with the overall process timeout, these timeouts are now
4932 treated as errors, giving rise to delivery failures by default. However,
4933 there is now a new Boolean option for the pipe transport called
4934 timeout_defer, which, if set TRUE, converts the failures into defers for
4935 both kinds of timeout. A transport filter timeout is now identified in
4938 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4939 systems where "make" and "gmake" are different, calling "gmake" at top
4940 level broke things. I've arranged for the value of $(MAKE) to be passed
4941 from the Makefile to this script so that it can call the same version of
4945 A note about Exim versions 4.44 and 4.50
4946 ----------------------------------------
4948 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4949 changes of various kinds. As a consequence, a big documentation update was
4950 needed. This delayed the release for rather longer than seemed good, especially
4951 in the light of a couple of (minor) security issues. Therefore, the changes
4952 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4953 release. So 4.44 and 4.50 are in effect two different branches that both start
4956 I have left the 4.50 change log unchanged; it contains all the changes since
4957 4.43. The change log for 4.44 is below; many of its items are identical to
4958 those for 4.50. This seems to be the most sensible way to preserve the
4959 historical information.
4965 1. Minor wording change to the doc/README.SIEVE file.
4967 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4968 computation of the current number of files was incorrect.
4970 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4971 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4974 4. Give more explanation in the error message when the command for a transport
4975 filter fails to execute.
4977 5. There are several places where Exim runs a non-Exim command in a
4978 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4979 was being done only for the command run by the queryprogram router. It is
4980 now done for all such subprocesses. The other cases are: ${run, transport
4981 filters, and the commands run by the lmtp and pipe transports.
4983 6. Added CONFIGURE_GROUP build-time option.
4985 7. Some older OS have a limit of 256 on the maximum number of file
4986 descriptors. Exim was using setrlimit() to set 1000 as a large value
4987 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4988 systems. I've change it so that if it can't get 1000, it tries for 256.
4990 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4991 was an oversight, and furthermore, ever since the addition of extra
4992 controls (e.g. 4.43/32), the checks on when to allow different forms of
4993 "control" were broken. There should now be diagnostics for all cases when a
4994 control that does not make sense is encountered.
4996 9. Added the /retain_sender option to "control=submission".
4998 10. $recipients is now available in the predata ACL (oversight).
5000 11. Tidy the search cache before the fork to do a delivery from a message
5001 received from the command line. Otherwise the child will trigger a lookup
5002 failure and thereby defer the delivery if it tries to use (for example) a
5003 cached ldap connection that the parent has called unbind on.
5005 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
5006 of $address_data from the recipient verification was clobbered by the
5007 sender verification.
5009 13. The value of address_data from a sender verification is now available in
5010 $sender_address_data in subsequent conditions in the ACL statement.
5012 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
5014 15. Added a new option "connect=<time>" to callout options, to set a different
5017 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
5018 was its contents. (It was OK if the option was not defined at all.)
5020 17. A "Completed" log line is now written for messages that are removed from
5021 the spool by the -Mrm option.
5023 18. New variables $sender_verify_failure and $recipient_verify_failure contain
5024 information about exactly what failed.
5026 19. Added -dd to debug only the daemon process.
5028 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
5029 handles timeouts, both on the server side and network timeouts. Renamed the
5030 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
5032 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
5033 It is now set to "smtps".
5035 22. $host_address is now set to the target address during the checking of
5036 ignore_target_hosts.
5038 23. When checking ignore_target_hosts for an ipliteral router, no host name was
5039 being passed; this would have caused $sender_host_name to have been used if
5040 matching the list had actually called for a host name (not very likely,
5041 since this list is usually IP addresses). A host name is now passed as
5044 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5045 code that specifies a non-restarting handler (typically sigaction() in
5046 modern systems) in an attempt to fix a rare and obscure crash bug.
5048 25. Narrowed the window for a race in the daemon that could cause it to ignore
5049 SIGCHLD signals. This is not a major problem, because they are used only to
5050 wake it up if nothing else does.
5052 26. A malformed maildirsize file could cause Exim to calculate negative values
5053 for the mailbox size or file count. Odd effects could occur as a result.
5054 The maildirsize information is now recalculated if the size or filecount
5057 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5058 support for a long time. Removed HAVE_SYS_VFS_H.
5060 28. Installed the latest version of exipick from John Jetmore.
5062 29. In an address list, if the pattern was not a regular expression, an empty
5063 subject address (from a bounce message) matched only if the pattern was an
5064 empty string. Non-empty patterns were not even tested. This was the wrong
5065 because it is perfectly reasonable to use an empty address as part of a
5066 database query. An empty address is now tested by patterns that are
5067 lookups. However, all the other forms of pattern expect the subject to
5068 contain a local part and a domain, and therefore, for them, an empty
5069 address still always fails if the pattern is not itself empty.
5071 30. Exim went into a mad DNS loop when attempting to do a callout where the
5072 host was specified on an smtp transport, and looking it up yielded more
5073 than one IP address.
5075 31. Re-factored the code for checking spool and log partition space into a
5076 function that finds that data and another that does the check. The former
5077 is then used to implement four new variables: $spool_space, $log_space,
5078 $spool_inodes, and $log_inodes.
5080 32. The RFC2047 encoding function was originally intended for short strings
5081 such as real names; it was not keeping to the 75-character limit for
5082 encoded words that the RFC imposes. It now respects the limit, and
5083 generates multiple encoded words if necessary. To be on the safe side, I
5084 have increased the buffer size for the ${rfc2047: expansion operator from
5087 33. It is now permitted to omit both strings after an "if" condition; if the
5088 condition is true, the result is "true". As before, when the second string
5089 is omitted, a false condition yields an empty string. This makes it less
5090 cumbersome to write custom ACL and router conditions.
5092 34. Failure to deliver a bounce message always caused it to be frozen, even if
5093 there was an errors_to setting on the router. The errors_to setting is now
5096 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5097 canonical form (fully expanded) before being placed in
5098 $sender_host_address.
5100 36. The table in the code that translates DNS record types into text (T_A to
5101 "A" for instance) was missing entries for NS and CNAME. It is just possible
5102 that this could have caused confusion if both these types were looked up
5103 for the same domain, because the text type is used as part of Exim's
5104 per-process caching. But the chance of anyone hitting this buglet seems
5107 37. The dnsdb lookup has been extended in a number of ways.
5109 (1) There is a new type, "zns", which walks up the domain tree until it
5110 finds some nameserver records. It should be used with care.
5112 (2) There is a new type, "mxh", which is like "mx" except that it returns
5113 just the host names, not the priorities.
5115 (3) It is now possible to give a list of domains (or IP addresses) to be
5116 looked up. The behaviour when one of the lookups defers can be
5117 controlled by a keyword.
5119 (4) It is now possible to specify the separator character for use when
5120 multiple records are returned.
5122 38. The dnslists ACL condition has been extended: it is now possible to supply
5123 a list of IP addresses and/or domains to be looked up in a particular DNS
5126 39. Added log_selector=+queue_time_overall.
5128 40. When running the queue in the test harness, wait just a tad after forking a
5129 delivery process, to get repeatability of debugging output.
5131 41. Include certificate and key file names in error message when GnuTLS fails
5132 to set them up, because the GnuTLS error message doesn't include the name
5133 of the failing file when there is a problem reading it.
5135 42. Allow both -bf and -bF in the same test run.
5137 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
5139 44. The "Exiscan patch" is now merged into the mainline Exim source.
5141 45. Sometimes the final signoff response after QUIT could fail to get
5142 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5143 < 0 before doing a fflush(). This bug looks as though it goes back to the
5144 introduction of TLS in release 3.20, but "sometimes" must have been rare
5145 because the tests only now provoked it.
5147 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
5148 (this can affect the format of dates).
5150 47. exim_tidydb, when checking for the continued existence of a message for
5151 which it has found a message-specific retry record, was not finding
5152 messages that were in split spool directories. Consequently, it was
5153 deleting retry records that should have stayed in existence.
5155 48. Steve fixed some bugs in eximstats.
5157 49. The SPA authentication driver was not abandoning authentication and moving
5158 on to the next authenticator when an expansion was forced to fail,
5159 contradicting the general specification for all authenticators. Instead it
5160 was generating a temporary error. It now behaves as specified.
5162 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
5163 (the order specifies the preference for clients). The order is now AES256,
5164 AES128, 3DES, ARCFOUR128.
5166 51. Small patch to Sieve code - explicitly set From: when generating an
5169 52. Exim crashed if a remote delivery caused a very long error message to be
5170 recorded - for instance if somebody sent an entire SpamAssassin report back
5171 as a large number of 550 error lines. This bug was coincidentally fixed by
5172 increasing the size of one of Exim's internal buffers (big_buffer) that
5173 happened as part of the Exiscan merge. However, to be on the safe side, I
5174 have made the code more robust (and fixed the comments that describe what
5177 53. Now that there can be additional text after "Completed" in log lines (if
5178 the queue_time_overall log selector is set), a one-byte patch to exigrep
5179 was needed to allow it to recognize "Completed" as not the last thing in
5182 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
5183 patch that reportedly fixes this has been added. I am not expert enough to
5184 create a test for it. This is what the patch creator wrote:
5186 "I found a little strange behaviour of ldap code when working with
5187 Windows 2003 AD Domain, where users was placed in more than one
5188 Organization Units. When I tried to give exim partial DN, the exit code
5189 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
5190 But simultaneously result of request was absolutely normal ldap result,
5191 so I produce this patch..."
5193 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
5194 REFERENCE, so I have modified the code to exclude the patch when that macro
5197 55. Some experimental protocols are using DNS PTR records for new purposes. The
5198 keys for these records are domain names, not reversed IP addresses. The
5199 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5200 leaves it alone. Component reversal etc. now happens only for IP addresses.
5203 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
5205 57. Double the size of the debug message buffer (to 2048) so that more of very
5206 long debug lines gets shown.
5208 58. The exicyclog utility now does better if the number of log files to keep
5209 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
5211 59. Two changes related to the smtp_active_hostname option:
5213 (1) $smtp_active_hostname is now available as a variable.
5214 (2) The default for smtp_banner uses $smtp_active_hostname instead
5215 of $primary_hostname.
5217 60. The host_aton() function is supposed to be passed a string that is known
5218 to be a valid IP address. However, in the case of IPv6 addresses, it was
5219 not checking this. This is a hostage to fortune. Exim now panics and dies
5220 if the condition is not met. A case was found where this could be provoked
5221 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5222 components; fortuitously, this particular loophole had already been fixed
5223 by change 4.50/55 above.
5225 If there are any other similar loopholes, the new check in host_aton()
5226 itself should stop them being exploited. The report I received stated that
5227 data on the command line could provoke the exploit when Exim was running as
5228 exim, but did not say which command line option was involved. All I could
5229 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5230 running as the user.
5233 61. There was a buffer overflow vulnerability in the SPA authentication code
5234 (which came originally from the Samba project). I have added a test to the
5235 spa_base64_to_bits() function which I hope fixes it.
5238 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
5239 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
5241 63. The daemon start-up calls getloadavg() while still root for those OS that
5242 need the first call to be done as root, but it missed one case: when
5243 deliver_queue_load_max is set with deliver_drop_privilege. This is
5244 necessary for the benefit of the queue runner, because there is no re-exec
5245 when deliver_drop_privilege is set.
5247 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
5248 This has been fixed.
5250 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5251 was in use, was not putting the data itself into the right store pool;
5252 consequently, it could be overwritten for a subsequent message in the same
5253 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5256 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
5258 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
5259 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
5260 the function were treating the return as a boolean value, which happened to
5261 work because 0=false and not-0=true, but is not correct code.
5263 68. The host_aton() function was not handling scoped IPv6 addresses (those
5264 with, for example, "%eth0" on the end) correctly.
5266 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
5267 negated items (that is, ~something) in unsigned ints. Some compilers
5268 apparently mutter when there is no cast.
5270 70. If an address verification called from an ACL failed, and did not produce a
5271 user-specific message (i.e. there was only a "system" message), nothing was
5272 put in $acl_verify_message. In this situation, it now puts the system
5275 71. Change 4.23/11 added synchronization checking at the start of an SMTP
5276 session; change 4.31/43 added the unwanted input to the log line - except
5277 that it did not do this in the start of session case. It now does.
5279 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
5280 This is wrong and can cause the other end to generate a synchronization
5281 error if it is another Exim or anything else that does the synchronization
5282 check. A QUIT command is no longer sent after a timeout.
5284 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
5285 during host lookups.
5287 74. The defer_ok option of callout verification was not working if it was used
5288 when verifying addresses in header lines, that is, for this case:
5290 verify = header_sender/callout=defer_ok
5292 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
5293 those file descriptors could be used for SMTP connections. If anything
5294 wrote to stderr (the example that came up was "warn" in embedded Perl), it
5295 could be sent to the SMTP client, causing chaos. The daemon now opens
5296 stdin, stdout, and stderr to /dev/null when it puts itself into the
5299 76. Arrange for output from Perl's "warn" command to be written to Exim's main
5300 log by default. The user can override this with suitable Perl magic.
5302 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
5303 the log message when discard triggers, was not working for the DATA ACL or
5304 for the non-SMTP ACL.
5306 78. Error message wording change in sieve.c.
5308 79. If smtp_accept_max_per_host was set, the number of connections could be
5309 restricted to fewer than expected, because the daemon was trying to set up
5310 a new connection before checking whether the processes handling previous
5311 connections had finished. The check for completed processes is now done
5312 earlier. On busy systems, this bug wouldn't be noticed because something
5313 else would have woken the daemon, and it would have reaped the completed
5316 80. If a message was submitted locally by a user whose login name contained one
5317 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
5318 It caused a spool format error. I have fixed the spool reading code. A
5319 related problem was that the "from" clause in the Received: line became
5320 illegal because of the space(s). It is now covered by ${quote_local_part.
5322 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
5325 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
5327 83. Patch from Sieve maintainer to fix unterminated string problem in
5328 "vacation" handling.
5330 84. Some minor changes to the Linux configuration files to help with other
5331 OS variants using glibc.
5333 85. One more patch for Sieve to update vacation handling to latest spec.
5336 ----------------------------------------------------
5337 See the note above about the 4.44 and 4.50 releases.
5338 ----------------------------------------------------
5344 1. Change 4.43/35 introduced a bug that caused file counts to be
5345 incorrectly computed when quota_filecount was set in an appendfile
5348 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
5349 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
5352 3. Give more explanation in the error message when the command for a transport
5353 filter fails to execute.
5355 4. There are several places where Exim runs a non-Exim command in a
5356 subprocess. The SIGUSR1 signal should be disabled for these processes. This
5357 was being done only for the command run by the queryprogram router. It is
5358 now done for all such subprocesses. The other cases are: ${run, transport
5359 filters, and the commands run by the lmtp and pipe transports.
5361 5. Some older OS have a limit of 256 on the maximum number of file
5362 descriptors. Exim was using setrlimit() to set 1000 as a large value
5363 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
5364 systems. I've change it so that if it can't get 1000, it tries for 256.
5366 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
5367 was an oversight, and furthermore, ever since the addition of extra
5368 controls (e.g. 4.43/32), the checks on when to allow different forms of
5369 "control" were broken. There should now be diagnostics for all cases when a
5370 control that does not make sense is encountered.
5372 7. $recipients is now available in the predata ACL (oversight).
5374 8. Tidy the search cache before the fork to do a delivery from a message
5375 received from the command line. Otherwise the child will trigger a lookup
5376 failure and thereby defer the delivery if it tries to use (for example) a
5377 cached ldap connection that the parent has called unbind on.
5379 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
5380 of $address_data from the recipient verification was clobbered by the
5381 sender verification.
5383 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
5384 was its contents. (It was OK if the option was not defined at all.)
5386 11. A "Completed" log line is now written for messages that are removed from
5387 the spool by the -Mrm option.
5389 12. $host_address is now set to the target address during the checking of
5390 ignore_target_hosts.
5392 13. When checking ignore_target_hosts for an ipliteral router, no host name was
5393 being passed; this would have caused $sender_host_name to have been used if
5394 matching the list had actually called for a host name (not very likely,
5395 since this list is usually IP addresses). A host name is now passed as
5398 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5399 code that specifies a non-restarting handler (typically sigaction() in
5400 modern systems) in an attempt to fix a rare and obscure crash bug.
5402 15. Narrowed the window for a race in the daemon that could cause it to ignore
5403 SIGCHLD signals. This is not a major problem, because they are used only to
5404 wake it up if nothing else does.
5406 16. A malformed maildirsize file could cause Exim to calculate negative values
5407 for the mailbox size or file count. Odd effects could occur as a result.
5408 The maildirsize information is now recalculated if the size or filecount
5411 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5412 support for a long time. Removed HAVE_SYS_VFS_H.
5414 18. Updated exipick to current release from John Jetmore.
5416 19. Allow an empty sender to be matched against a lookup in an address list.
5417 Previously the only cases considered were a regular expression, or an
5420 20. Exim went into a mad DNS lookup loop when doing a callout where the
5421 host was specified on the transport, if the DNS lookup yielded more than
5424 21. The RFC2047 encoding function was originally intended for short strings
5425 such as real names; it was not keeping to the 75-character limit for
5426 encoded words that the RFC imposes. It now respects the limit, and
5427 generates multiple encoded words if necessary. To be on the safe side, I
5428 have increased the buffer size for the ${rfc2047: expansion operator from
5431 22. Failure to deliver a bounce message always caused it to be frozen, even if
5432 there was an errors_to setting on the router. The errors_to setting is now
5435 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5436 canonical form (fully expanded) before being placed in
5437 $sender_host_address.
5439 24. Updated eximstats to version 1.33
5441 25. Include certificate and key file names in error message when GnuTLS fails
5442 to set them up, because the GnuTLS error message doesn't include the name
5443 of the failing file when there is a problem reading it.
5445 26. Expand error message when OpenSSL has problems setting up cert/key files.
5448 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
5449 (this can affect the format of dates).
5451 28. exim_tidydb, when checking for the continued existence of a message for
5452 which it has found a message-specific retry record, was not finding
5453 messages that were in split spool directories. Consequently, it was
5454 deleting retry records that should have stayed in existence.
5456 29. eximstats updated to version 1.35
5457 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
5458 1.35 - bugfix such that pie charts by volume are generated correctly
5460 30. The SPA authentication driver was not abandoning authentication and moving
5461 on to the next authenticator when an expansion was forced to fail,
5462 contradicting the general specification for all authenticators. Instead it
5463 was generating a temporary error. It now behaves as specified.
5465 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
5466 (the order specifies the preference for clients). The order is now AES256,
5467 AES128, 3DES, ARCFOUR128.
5469 31. Small patch to Sieve code - explicitly set From: when generating an
5472 32. Exim crashed if a remote delivery caused a very long error message to be
5473 recorded - for instance if somebody sent an entire SpamAssassin report back
5474 as a large number of 550 error lines. This bug was coincidentally fixed by
5475 increasing the size of one of Exim's internal buffers (big_buffer) that
5476 happened as part of the Exiscan merge. However, to be on the safe side, I
5477 have made the code more robust (and fixed the comments that describe what
5480 33. Some experimental protocols are using DNS PTR records for new purposes. The
5481 keys for these records are domain names, not reversed IP addresses. The
5482 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5483 leaves it alone. Component reversal etc. now happens only for IP addresses.
5486 34. The host_aton() function is supposed to be passed a string that is known
5487 to be a valid IP address. However, in the case of IPv6 addresses, it was
5488 not checking this. This is a hostage to fortune. Exim now panics and dies
5489 if the condition is not met. A case was found where this could be provoked
5490 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5491 components; fortuitously, this particular loophole had already been fixed
5492 by change 4.50/55 or 4.44/33 above.
5494 If there are any other similar loopholes, the new check in host_aton()
5495 itself should stop them being exploited. The report I received stated that
5496 data on the command line could provoke the exploit when Exim was running as
5497 exim, but did not say which command line option was involved. All I could
5498 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5499 running as the user.
5502 35. There was a buffer overflow vulnerability in the SPA authentication code
5503 (which came originally from the Samba project). I have added a test to the
5504 spa_base64_to_bits() function which I hope fixes it.
5507 36. The daemon start-up calls getloadavg() while still root for those OS that
5508 need the first call to be done as root, but it missed one case: when
5509 deliver_queue_load_max is set with deliver_drop_privilege. This is
5510 necessary for the benefit of the queue runner, because there is no re-exec
5511 when deliver_drop_privilege is set.
5513 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5514 was in use, was not putting the data itself into the right store pool;
5515 consequently, it could be overwritten for a subsequent message in the same
5516 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5519 38. Sometimes the final signoff response after QUIT could fail to get
5520 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5521 < 0 before doing a fflush(). This bug looks as though it goes back to the
5522 introduction of TLS in release 3.20, but "sometimes" must have been rare
5523 because the tests only now provoked it.
5529 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5530 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5531 Now it returns an integer. A number of calls were still expecting a T/F
5532 return. Fortuitously, in all cases, the tests worked in OK situations,
5533 which is the norm. However, things would have gone wrong on any write
5534 failures on the smtp file descriptor. This function is used when sending
5535 messages over SMTP and also when doing verify callouts.
5537 2. When Exim is called to do synchronous delivery of a locally submitted
5538 message (the -odf or -odi options), it no longer closes stderr before doing
5541 3. Implemented the mua_wrapper option.
5543 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5545 5. Implemented the functions header_remove(), header_testname(),
5546 header_add_at_position(), and receive_remove_recipient(), and exported them
5549 6. If an ACL "warn" statement specified the addition of headers, Exim already
5550 inserted X-ACL-Warn: at the start if there was no header name. However, it
5551 was not making this test for the second and subsequent header lines if
5552 there were newlines in the string. This meant that an invalid header could
5553 be inserted if Exim was badly configured.
5555 7. Allow an ACL "warn" statement to add header lines at the start or after all
5556 the Received: headers, as well as at the end.
5558 8. Added the rcpt_4xx retry error code.
5560 9. Added postmaster_mailfrom=xxx to callout verification option.
5562 10. Added mailfrom=xxxx to the callout verification option, for verify=
5565 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5566 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5568 12. Inserted some casts to stop certain compilers warning when using pointer
5569 differences as field lengths or precisions in printf-type calls (mostly
5570 affecting debugging statements).
5572 13. Added optional readline() support for -be (dynamically loaded).
5574 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5575 same clock tick as a message's arrival, so that its received time was the
5576 same as the "first fail" time on the retry record, and that message
5577 remained on the queue past the ultimate address timeout, every queue runner
5578 would try a delivery (because it was past the ultimate address timeout) but
5579 after another failure, the ultimate address timeout, which should have then
5580 bounced the address, did not kick in. This was a "< instead of <=" error;
5581 in most cases the first failure would have been in the next clock tick
5582 after the received time, and all would be well.
5584 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5585 being recognized when the domain list was tested by the match_domain
5586 condition in an expansion string.
5588 16. Added the ${str2b64: operator.
5590 17. Exim was always calling setrlimit() to set a large limit for the number of
5591 processes, without checking whether the existing limit was already
5592 adequate. (It did check for the limit on file descriptors.) Furthermore,
5593 errors from getrlimit() and setrlimit() were being ignored. Now they are
5594 logged to the main and panic logs, but Exim does carry on, to try to do its
5595 job under whatever limits there are.
5597 18. Imported PCRE 5.0.
5599 19. Trivial typo in log message " temporarily refused connection" (the leading
5602 20. If the log selector return_path_on_delivery was set and an address was
5603 redirected to /dev/null, the delivery process crashed because it assumed
5604 that a return path would always be set for a "successful" delivery. In this
5605 case, the whole delivery is bypassed as an optimization, and therefore no
5608 21. Internal re-arrangement: the function for sending a challenge and reading
5609 a response while authentication was assuming a zero-terminated challenge
5610 string. It's now changed to take a pointer and a length, to allow for
5611 binary data in such strings.
5613 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5615 23. Exim was not respecting finduser_retries when seeking the login of the
5616 uid under which it was called; it was always trying 10 times. (The default
5617 setting of finduser_retries is zero.) Also, it was sleeping after the final
5618 failure, which is pointless.
5620 24. Implemented tls_on_connect_ports.
5622 25. Implemented acl_smtp_predata.
5624 26. If the domain in control=submission is set empty, Exim assumes that the
5625 authenticated id is a complete email address when it generates From: or
5626 Sender: header lines.
5628 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5629 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5630 chown and chgrp in /bin and hostname in /usr/bin.
5632 28. Exim was keeping the "process log" file open after each use, just as it
5633 does for the main log. This opens the possibility of it remaining open for
5634 long periods when the USR1 signal hits a daemon. Occasional processlog
5635 errors were reported, that could have been caused by this. Anyway, it seems
5636 much more sensible not to leave this file open at all, so that is what now
5639 29. The long-running daemon process does not normally write to the log once it
5640 has entered its main loop, and it closes the log before doing so. This is
5641 so that log files can straightforwardly be renamed and moved. However,
5642 there are a couple of unusual error situations where the daemon does write
5643 log entries, and I had neglected to close the log afterwards.
5645 30. The text of an SMTP error response that was received during a remote
5646 delivery was being truncated at 512 bytes. This is too short for some of
5647 the long messages that one sometimes sees. I've increased the limit to
5650 31. It is now possible to make retry rules that apply only when a message has a
5651 specific sender, in particular, an empty sender.
5653 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5654 it possible to be selective about when SMTP synchronization is enforced.
5656 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5658 32. Implemented hosts_connection_nolog.
5660 33. Added an ACL for QUIT.
5662 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5665 35. Added mailbox_size and mailbox_filecount to appendfile.
5667 36. Added control = no_multiline_responses to ACLs.
5669 37. There was a bug in the logic of the code that waits for the clock to tick
5670 in the case where the clock went backwards by a substantial amount such
5671 that the microsecond fraction of "now" was more than the microsecond
5672 fraction of "then" (but the whole seconds number was less).
5674 38. Added support for the libradius Radius client library this is found on
5675 FreeBSD (previously only the radiusclient library was supported).
5681 1. When certain lookups returned multiple values in the form name=value, the
5682 quoting of the values was not always being done properly. Specifically:
5683 (a) If the value started with a double quote, but contained no whitespace,
5685 (b) If the value contained whitespace other than a space character (i.e.
5686 tabs or newlines or carriage returns) it was not quoted.
5687 This fix has been applied to the mysql and pgsql lookups by writing a
5688 separate quoting function and calling it from the lookup code. The fix
5689 should probably also be applied to nisplus, ibase and oracle lookups, but
5690 since I cannot test any of those, I have not disturbed their existing code.
5692 2. A hit in the callout cache for a specific address caused a log line with no
5693 reason for rejecting RCPT. Now it says "Previous (cached) callout
5694 verification failure".
5696 3. There was an off-by-one bug in the queryprogram router. An over-long
5697 return line was truncated at 256 instead of 255 characters, thereby
5698 overflowing its buffer with the terminating zero. As well as fixing this, I
5699 have increased the buffer size to 1024 (and made a note to document this).
5701 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5702 when Exim is waiting for an SMTP response from a remote server, Exim
5703 restarts its select() call on the socket, thereby resetting its timeout.
5704 This is not a problem when such interrupts are rare. Somebody set up a cron
5705 job to run exiwhat every 2 minutes, which is less than the normal select()
5706 timeout (5 or 10 minutes). This meant that the select() timeout never
5707 kicked in because it was always reset. I have fixed this by comparing the
5708 time when an interrupt arrives with the time at the start of the first call
5709 to select(). If more time than the timeout has elapsed, the interrupt is
5710 treated as a timeout.
5712 5. Some internal re-factoring in preparation for the addition of Sieve
5713 extensions (by MH). In particular, the "personal" test is moved to a
5714 separate function, and given an option for scanning Cc: and Bcc: (which is
5715 not set for Exim filters).
5717 6. When Exim created an email address using the login of the caller as the
5718 local part (e.g. when creating a From: or Sender: header line), it was not
5719 quoting the local part when it contained special characters such as @.
5721 7. Installed new OpenBSD configuration files.
5723 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5724 try to make them clearer.
5726 9. Callout options, other than the timeout value, were being ignored when
5727 verifying sender addresses in header lines. For example, when using
5729 verify = header_sender/callout=no_cache
5731 the cache was (incorrectly) being used.
5733 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5734 only the Cygwin environment.
5736 11. When return_path_on_delivery was set as a log selector, if different remote
5737 addresses in the same message used different return paths and parallel
5738 remote delivery occurred, the wrong values would sometimes be logged.
5739 (Whenever a remote delivery process finished, the return path value from
5740 the most recently started remote delivery process was logged.)
5742 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5743 lines when AUTH and/or TLS are in use. This is the "received protocol"
5744 field. Exim used to use "asmtp" for authenticated SMTP, without any
5745 indication (in the protocol name) for TLS use. Now it follows the RFC and
5746 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5747 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5748 names appear in log lines as well as in Received: header lines.
5750 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5751 extensions, and comparison tests, and to fix some bugs.
5753 14. Changes to the "personal" filter test:
5755 (1) The test was buggy in that it was just doing the equivalent of
5756 "contains" tests on header lines. For example, if a user's address was
5757 anne@some.where, the "personal" test would incorrectly be true for
5759 To: susanne@some.where
5761 This test is now done by extracting each address from the header in turn,
5762 and checking the entire address. Other tests that are part of "personal"
5763 are now done using regular expressions (for example, to check local parts
5764 of addresses in From: header lines).
5766 (2) The list of non-personal local parts in From: addresses has been
5767 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5768 taken from the Sieve specification recommendations.
5770 (3) If the message contains any header line starting with "List-" it is
5771 treated as non-personal.
5773 (4) The test for "circular" in the Subject: header line has been removed
5774 because it now seems ill-conceived.
5776 15. Minor typos in src/EDITME comments corrected.
5778 16. Installed latest exipick from John Jetmore.
5780 17. If headers_add on a router specified a text string that was too long for
5781 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5782 of string_sprintf() is now avoided.
5784 18. $message_body_size was not set (it was always zero) when running the DATA
5785 ACL and the local_scan() function.
5787 19. For the "mail" command in an Exim filter, no default was being set for
5788 the once_repeat time, causing a random time value to be used if "once" was
5789 specified. (If the value happened to be <= 0, no repeat happened.) The
5790 default is now 0s, meaning "never repeat". The "vacation" command was OK
5791 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5792 (I found it when inspecting the code).
5794 20. There is now an overall timeout for performing a callout verification. It
5795 defaults to 4 times the callout timeout, which applies to individual SMTP
5796 commands during the callout. The overall timeout applies when there is more
5797 than one host that can be tried. The timeout is checked before trying the
5798 next host. This prevents very long delays if there are a large number of
5799 hosts and all are timing out (e.g. when the network connections are timing
5800 out). The value of the overall timeout can be changed by specifying an
5801 additional sub-option for "callout", called "maxwait". For example:
5803 verify = sender/callout=5s,maxwait=20s
5805 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5806 to the end before writing, but this should make it even safer).
5808 22. Exim was forgetting that it had advertised PIPELINING for the second and
5809 subsequent messages on an SMTP connection. It was also not resetting its
5810 memory on STARTTLS and an internal HELO.
5812 23. When Exim logs an SMTP synchronization error within a session, it now
5813 records whether PIPELINING has been advertised or not.
5815 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5816 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5817 rather than long int.
5819 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5821 26. Added the never_mail option to autoreply.
5827 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5828 crash if the getsockname() call failed; this can happen if a connection is
5829 closed very soon after it is established. The problem was simply in the
5830 order in which certain operations were done, causing Exim to try to write
5831 to the SMTP stream before it had set up the file descriptor. The bug has
5832 been fixed by making things happen in the correct order.
5838 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5839 before the connection was closed, thus losing the rejection response.
5841 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5842 some early Solaris releases, but causes trouble in current releases where
5843 socklen_t is defined.
5845 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5848 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5851 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5852 that releasing the top of it at the end releases what was used for sub-
5853 expansions (unless the block got too big). However, discard this block if
5854 the first thing is a variable or header, so that we can use its block when
5855 it is dynamic (useful for very large $message_headers, for example).
5857 6. Lookups now cache *every* query, not just the most recent. A new, separate
5858 store pool is used for this. It can be recovered when all lookup caches are
5859 flushed. Lookups now release memory at the end of their result strings.
5860 This has involved some general refactoring of the lookup sources.
5862 7. Some code has been added to the store_xxx() functions to reduce the amount
5863 of flapping under certain conditions.
5865 8. log_incoming_interface used to affect only the <= reception log lines. Now
5866 it causes the local interface and port to be added to several more SMTP log
5867 lines, for example "SMTP connection from", and rejection lines.
5869 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5871 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5873 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5874 could be overwritten at the end of the current message (or the start of a
5875 new message if it was set in a HELO ACL). The value is now preserved for
5876 the duration of the SMTP connection.
5878 12. If a transport had a headers_rewrite setting, and a matching header line
5879 contained an unqualified address, that address was qualified, even if it
5880 did not match any rewriting rules. The underlying bug was that the values
5881 of the flags that permit the existence of unqualified sender and recipient
5882 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5883 non-local messages, and by -bnq for local messages) were not being
5884 preserved with the message after it was received.
5886 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5887 "next input=" as part of the text comprising the host identity instead of
5888 the correct text. The code was using the same buffer for two different
5889 strings. However, depending on which order the printing function evaluated
5890 its arguments, the bug did not always show up. Under Linux, for example, my
5891 test suite worked just fine.
5893 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5894 doesn't work with some older versions of Perl. It has been changed to "my",
5895 which in any case is probably the better facility to use.
5897 15. A really picky compiler found some instances of statements for creating
5898 error messages that either had too many or two few arguments for the format
5901 16. The size of the buffer for calls to the DNS resolver has been increased
5902 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5903 for addresses that have a lot of PTR records. This alleviates a problem; it
5904 does not fully solve it.
5906 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5907 buffer now truncates the list and logs the incident, which is the same
5908 action as happens when Exim is looking up a host name and its aliases.
5909 Previously in this situation something unpredictable would happen;
5910 sometimes it was "internal error: store_reset failed".
5912 18. If a server dropped the connection unexpectedly when an Exim client was
5913 using GnuTLS and trying to read a response, the client delivery process
5914 crashed while trying to generate an error log message.
5916 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5917 string, for example:
5919 warn message = H1: something\nH2: something
5921 the text was added as a single header line from Exim's point of view
5922 though it ended up OK in the delivered message. However, searching for the
5923 second and subsequent header lines using $h_h2: did not work. This has been
5924 fixed. Similarly, if a system filter added multiple headers in this way,
5925 the routers could not see them.
5927 20. Expanded the error message when iplsearch is called with an invalid key to
5928 suggest using net-iplsearch in a host list.
5930 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5931 ACLs are no longer actually imposed (and a message to that effect is
5934 22. If a "gecos" field in a passwd entry contained escaped characters, in
5935 particular, if it contained a \" sequence, Exim got it wrong when building
5936 a From: or a Sender: header from that name. A second bug also caused
5937 incorrect handling when an unquoted " was present following a character
5938 that needed quoting.
5940 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5941 was not being matched caselessly.
5943 24. Arranged for all hyphens in the exim.8 source to be escaped with
5946 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5947 properly. Recipient callout cache records were still being keyed to include
5948 the sender, even when use_sender was set false. This led to far more
5949 callouts that were necessary. The sender is no longer included in the key
5950 when use_sender is false.
5952 26. Added "control = submission" modifier to ACLs.
5954 27. Added the ${base62d: operator to decode base 62 numbers.
5956 28. dnsdb lookups can now access SRV records.
5958 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5959 the configuration file.
5961 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5962 (-v) mode. This makes the output for a verbose queue run more intelligible.
5964 31. Added a use_postmaster feature to recipient callouts.
5966 32. Added the $body_zerocount variable, containing the number of binary zero
5967 bytes in the message body.
5969 33. The time of last modification of the "new" subdirectory is now used as the
5970 "mailbox time last read" when there is a quota error for a maildir
5973 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5975 35. Added +ignore_unknown as a special item in host lists.
5977 36. Code for decoding IPv6 addresses in host lists is now included, even if
5978 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5979 address was recognized as an IP address, but was then not correctly decoded
5980 into binary, causing unexpected and incorrect effects when compared with
5987 1. Very minor rewording of debugging text in manualroute to say "list of
5988 hosts" instead of "hostlist".
5990 2. If verify=header_syntax was set, and a header line with an unqualified
5991 address (no domain) and a large number of spaces between the end of the
5992 name and the colon was received, the reception process suffered a buffer
5993 overflow, and (when I tested it) crashed. This was caused by some obsolete
5994 code that should have been removed. The fix is to remove it!
5996 3. When running in the test harness, delay a bit after writing a bounce
5997 message to get a bit more predictability in the log output.
5999 4. Added a call to search_tidyup() just before forking a reception process. In
6000 theory, someone could use a lookup in the expansion of smtp_accept_max_
6001 per_host which, without the tidyup, could leave open a database connection.
6003 5. Added the variables $recipient_data and $sender_data which get set from a
6004 lookup success in an ACL "recipients" or "senders" condition, or a router
6005 "senders" option, similar to $domain_data and $local_part_data.
6007 6. Moved the writing of debug_print from before to after the "senders" test
6010 7. Change 4.31/66 (moving the time when the Received: is generated) caused
6011 problems for message scanning, either using a data ACL, or using
6012 local_scan() because the Received: header was not generated till after they
6013 were called (in order to set the time as the time of reception completion).
6014 I have revised the way this works. The header is now generated after the
6015 body is received, but before the ACL or local_scan() are called. After they
6016 are run, the timestamp in the header is updated.
6022 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
6023 before starting a queue runner without re-exec. This happened only when
6024 deliver_drop_privilege was set or when the Exim user was set to root. The
6025 effect of the bug was that timeouts during subsequent deliveries caused
6026 crashes instead of being properly handled. The handler is now left at its
6027 default (and expected) setting.
6029 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
6030 message, again when deliver_drop_privilege is set or Exim is run as root.
6031 The bug described in (1) was not present in this case, but the tidying up
6032 of the other signals was missing. I have made the two cases consistent.
6034 3. The ignore_target_hosts setting on a manualroute router was being ignored
6035 for hosts that were looked up using the /MX notation.
6037 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
6040 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
6041 operated on the sender address. After changing the $sender_address to <>
6042 for the sender address verify, Exim was re-instated it as the original
6043 (before rewriting) address, but remembering that it had rewritten it, so it
6044 wasn't rewriting it again. This bug also had the effect of breaking the
6045 sender address verification caching when the sender address was rewritten.
6047 6. The ignore_target_hosts option was being ignored by the ipliteral router.
6048 This has been changed so that if the ip literal address matches
6049 ignore_target_hosts, the router declines.
6051 7. Added expansion conditions match_domain, match_address, and match_local_
6052 part (NOT match_host).
6054 8. The placeholder for the Received: header didn't have a length field set.
6056 9. Added code to Exim itself and to exim_lock to test for a specific race
6057 condition that could lead to file corruption when using MBX delivery. The
6058 issue is with the lockfile that is created in /tmp. If this file is removed
6059 after a process has opened it but before that process has acquired a lock,
6060 there is the potential for a second process to recreate the file and also
6061 acquire a lock. This could lead to two Exim processes writing to the file
6062 at the same time. The added code performs the same test as UW imapd; it
6063 checks after acquiring the lock that its file descriptor still refers to
6064 the same named file.
6066 10. The buffer for building added header lines was of fixed size, 8192 bytes.
6067 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
6070 11. Added the smtp_active_hostname option. If used, this will typically be made
6071 to depend on the incoming interface address. Because $interface_address is
6072 not set up until the daemon has forked a reception process, error responses
6073 that can happen earlier (such as "too many connections") no longer contain
6076 12. If an expansion in a condition on a "warn" statement fails because a lookup
6077 defers, the "warn" statement is abandoned, and the next ACL statement is
6078 processed. Previously this caused the whole ACL to be aborted.
6080 13. Added the iplsearch lookup type.
6082 14. Added ident_timeout as a log selector.
6084 15. Added tls_certificate_verified as a log selector.
6086 16. Added a global option tls_require_ciphers (compare the smtp transport
6087 option of the same name). This controls incoming TLS connections.
6089 17. I finally figured out how to make tls_require_ciphers do a similar thing
6090 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
6091 before starting the TLS session.
6093 18. Tabs are now shown as \t in -bP output.
6095 19. If the log selector return_path_on_delivery was set, Exim crashed when
6096 bouncing a message because it had too many Received: header lines.
6098 20. If two routers both had headers_remove settings, and the first one included
6099 a superfluous trailing colon, the final name in the first list and the
6100 first name in the second list were incorrectly joined into one item (with a
6101 colon in the middle).
6107 1. Added -C and -D options to the exinext utility, mainly to make it easier
6108 to include in the automated testing, but these could be helpful when
6109 multiple configurations are in use.
6111 2. The exinext utility was not formatting the output nicely when there was
6112 an alternate port involved in the retry record key, nor when there was a
6113 message id as well (for retries that were specific to a specific message
6114 and a specific host). It was also confused by IPv6 addresses, because of
6115 the additional colons they contain. I have fixed the IPv4 problem, and
6116 patched it up to do a reasonable job for IPv6.
6118 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
6119 delivery, the log line now contains "pipelined" if PIPELINING was used.
6121 4. An SMTP transport process used to panic and die if the bind() call to set
6122 an explicit outgoing interface failed. This has been changed; it is now
6123 treated in the same way as a connect() failure.
6125 5. A reference to $sender_host_name in the part of a conditional expansion
6126 that was being skipped was still causing a DNS lookup. This no longer
6129 6. The def: expansion condition was not recognizing references to header lines
6130 that used bh_ and bheader_.
6132 7. Added the _cache feature to named lists.
6134 8. The code for checking quota_filecount in the appendfile transport was
6135 allowing one more file than it should have been.
6137 9. For compatibility with Sendmail, the command line option
6145 and sets the incoming protocol and host name (for trusted callers). The
6146 host name and its colon can be omitted when only the protocol is to be set.
6147 Note the Exim already has two private options, -pd and -ps, that refer to
6148 embedded Perl. It is therefore impossible to set a protocol value of "d" or
6149 "s", but I don't think that's a major issue.
6151 10. A number of refactoring changes to the code, none of which should affect
6154 (a) The number of logging options was getting close to filling up the
6155 32-bit word that was used as a bit map. I have split them into two classes:
6156 those that are passed in the argument to log_write(), and those that are
6157 only ever tested independently outside of that function. These are now in
6158 separate 32-bit words, so there is plenty of room for expansion again.
6159 There is no change in the user interface or the logging behaviour.
6161 (b) When building, for example, log lines, the code previously used a
6162 macro that called string_cat() twice, in order to add two strings. This is
6163 not really sufficiently general. Furthermore, there was one instance where
6164 it was actually wrong because one of the argument was used twice, and in
6165 one call a function was used. (As it happened, calling the function twice
6166 did not affect the overall behaviour.) The macro has been replaced by a
6167 function that can join an arbitrary number of extra strings onto a growing
6170 (c) The code for expansion conditions now uses a table and a binary chop
6171 instead of a serial search (which was left over from when there were very
6172 few conditions). Also, it now recognizes conditions like "pam" even when
6173 the relevant support is not compiled in: a suitably worded error message is
6174 given if an attempt is made to use such a condition.
6176 11. Added ${time_interval:xxxxx}.
6178 12. A bug was causing one of the ddress fields not to be passed back correctly
6179 from remote delivery subprocesses. The field in question was not being
6180 subsequently used, so this caused to problems in practice.
6182 13. Added new log selectors queue_time and deliver_time.
6184 14. Might have fixed a bug in maildirsizefile handling that threw up
6185 "unexpected character" debug warnings, and recalculated the data
6186 unnecessarily. In any case, I expanded the warning message to give more
6189 15. Added the message "Restricted characters in address" to the statements in
6190 the default ACL that block characters like @ and % in local parts.
6192 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
6193 Three changes have been made:
6195 (a) There was a serious bug; a negative response to MAIL caused the whole
6196 recipient domain to be cached as invalid, thereby blocking all messages
6197 to all local parts at the same domain, from all senders. This bug has
6198 been fixed. The domain is no longer cached after a negative response to
6199 MAIL if the sender used is not empty.
6201 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
6204 (c) A new callout option, "use_sender" has been added for people who want
6205 the modified behaviour.
6211 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
6214 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
6215 indeed breaks things for older releases.
6217 3. Added additional logging to the case where there is a problem reading data
6218 from a filter that is running in a subprocess using a pipe, in order to
6219 try to track down a specific problem.
6221 4. Testing facility fudge: when running in the test harness and attempting
6222 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
6223 getting "No route to host". Convert this to a timeout.
6225 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
6228 6. Some OS don't have socklen_t but use size_t instead. This affects the
6229 fifth argument of getsockopt() amongst other things. This is now
6230 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
6231 can be set for individual OS. I have set it for SunOS5, OSF1, and
6232 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
6233 some earlier ones do not.
6235 7. Change 4.30/15 was not doing the test caselessly.
6237 8. The standard form for an IPv6 address literal was being rejected by address
6238 parsing in, for example, MAIL and RCPT commands. An example of this kind of
6239 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6240 this, as well as the form without the "IPv6" on the front (but only when
6241 address literals are enabled, of course).
6243 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
6245 10. Exim crashed if a message with an empty sender address specified by -f
6246 encountered a router with an errors_to setting. This could be provoked only
6247 by a command such as
6251 where an empty string was supplied; "<>" did not hit this bug.
6253 11. Installed PCRE release 4.5.
6255 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
6256 remained set. It is now erased.
6258 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
6259 times from message ids (which are base 36 rather than the normal 62).
6261 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
6262 were being counted as actual protocol errors, and logged if the log
6263 selector +smtp_protocol_error was set. One cannot be perfect in this test,
6264 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
6265 and DATA following a set of rejected RCPTs do not count as protocol errors.
6266 In other words, Exim assumes they were pipelined, though this may not
6267 actually be the case. Of course, in all cases the client gets an
6268 appropriate error code.
6270 15. If a lookup fails in an ACL condition, a message about the failure may
6271 be available; it is used if testing the ACL cannot continue, because most
6272 such messages specify what the cause of the deferral is. However, some
6273 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
6274 that caused an old message to be retained and used if a later statement
6275 caused a defer, replacing the real cause of the deferral.
6277 16. If an IP address had so many PTR records that the DNS lookup buffer
6278 was not large enough to hold them, Exim could crash while trying to process
6279 the truncated data. It now detects and logs this case.
6281 17. Further to 4.21/58, another change has been made: if (and only if) the
6282 first line of a message (the first header line) ends with CRLF, a bare LF
6283 in a subsequent header line has a space inserted after it, so as not to
6284 terminate the header.
6286 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
6287 unnecessarily, used atoi() instead of strtol(), and didn't check the
6288 termination when getting file sizes from file names by regex.
6290 19. Completely re-implemented the support for maildirsize files, in the light
6291 of a number of problems with the previous contributed implementation
6292 (4.30/29). In particular:
6294 . If the quota is zero, the maildirsize file is maintained, but no quota is
6297 . If the maildir directory does not exist, it is created before any attempt
6298 to write a maildirsize file.
6300 . The quota value in the file is just a cache; if the quota is changed in
6301 the transport, the new value overrides.
6303 . A regular expression is available for excluding directories from the
6306 20. The autoreply transport checks the characters in options that define the
6307 message's headers; it allows continued headers, but it was checking with
6308 isspace() after an embedded newline instead of explicitly looking for a
6311 21. If all the "regular" hosts to which an address was routed had passed their
6312 expiry times, and had not reached their retry times, the address was
6313 bounced, even if fallback hosts were defined. Now Exim should go on to try
6316 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
6317 equivalent code in the SMTP transport. Some hosts send humungous responses
6318 to HELO/EHLO, more than 1024 it seems.
6320 23. Refactoring: code in filter.c used (void *) for "any old type" but this
6321 gives compiler warnings in some environments. I've now done it "properly",
6324 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
6325 (because of problems with the built-in one) was declared to return uschar *
6326 instead of char *, causing compiler failure.
6328 25. Fixed a file descriptor leak when processing alias/forward files.
6330 26. Fixed a minor format string issue in dbfn.c.
6332 27. Typo in exim.c: ("dmbnz" for "dbmnz").
6334 28. If a filter file refered to $h_xxx or $message_headers, and the headers
6335 contained RFC 2047 "words", Exim's memory could, under certain conditions,
6338 29. When a sender address is verified, it is cached, to save repeating the test
6339 when there is more than one recipient in a message. However, when the
6340 verification involves a callout, it is possible for different callout
6341 options to be set for different recipients. It is too complicated to keep
6342 track of this in the cache, so now Exim always runs a verification when a
6343 callout is required, relying on the callout cache for the optimization.
6344 The overhead is duplication of the address routing, but this should not be
6347 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
6348 to be verified with callout=postmaster, and the main callout worked but the
6349 postmaster check failed, the verification correctly failed. However, if a
6350 subsequent RCPT command asked for sender verification *without* the
6351 postmaster check, incorrect caching caused this verification also to fail,
6354 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
6355 it was not caching the DNS options (qualify_single, search_parents) that
6356 were used when the lookup failed. A subsequent lookup with different
6357 options therefore always gave the same answer, though there were cases
6358 where it should not have. (Example: a "domains = !$mx_any" option on a
6359 dnslookup router: the "domains" option is always processed without any
6360 widening, but the router might have qualify_single set.) Now Exim uses the
6361 cached value only when the same options are set.
6363 32. Added John Jetmore's "exipick" utility to the distribution.
6365 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
6366 than a timeout (e.g. a certificate is required, and is not provided), an
6367 Exim server now closes the connection immediately. Previously it waited for
6368 the client to close - but if the client is SSL, it seems that they each
6369 wait for each other, leading to a delay before one of them times out.
6371 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
6372 maintained 0.8.x compatibility because I don't think many are using it, and
6373 it is clearly obsolete.
6375 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
6378 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
6379 client certificate was expired. A simple patch fixes this, though I don't
6380 understand the full logic of why the verify callback is called multiple
6383 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
6384 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
6385 which causes problems with some clients (such as the Certicom SSL Plus
6386 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
6387 disables the coutermeasure allowing Eudora to connect."
6389 38. Exim was not checking that a write() to a log file succeeded. This could
6390 lead to Bad Things if a log got too big, in particular if it hit a file
6391 size limit. Exim now panics and dies if it cannot write to a log file, just
6392 as it does if it cannot open a log file.
6394 39. Modified OS/Makefile-Linux so that it now contains
6396 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
6398 The two -D definitions ensure that Exim is compiled with large file
6399 support, which makes it possible to handle log files that are bigger than
6402 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
6403 instance) a domain was checked against a named list that involved a lookup,
6404 causing $domain_data to be set, then another domain was checked against the
6405 same list, then the first domain was re-checked, the value of $domain_data
6406 after the final check could be wrong. In particular, if the second check
6407 failed, it could be set empty. This bug probably also applied to
6410 41. The strip_trailing_dot option was not being applied to the address given
6411 with the -f command-line option.
6413 42. The code for reading a message's header from the spool was incrementing
6414 $received_count, but never initializing it. This meant that the value was
6415 incorrect (doubled) while delivering a message in the same process in which
6416 it was received. In the most common configuration of Exim, this never
6417 happens - a fresh exec is done - but it can happen when
6418 deliver_drop_privilege is set.
6420 43. When Exim logs an SMTP synchronization error - client data sent too soon -
6421 it now includes up to 150 characters of the unexpected data in the log
6424 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
6425 and building data strings. The size of both of these buffers was 10 000
6426 bytes - far larger than anybody would *ever* want, thought I. Needless to
6427 say, somebody hit the limit. I have increased the maximum line length to
6428 20 000 and the maximum data length of concatenated lines to 100 000. I have
6429 also fixed two bugs, because there was no checking on these buffers. Tsk,
6430 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
6431 buffer is too small.
6433 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
6434 lsearch lookups. Now it does.
6436 46. When parsing a route_list item in a manualroute router, a fixed-length
6437 buffer was used for the list of hosts. I made this 1024 bytes long,
6438 thinking that nobody would ever have a list of hosts that long. Wrong.
6439 Somebody had a whole pile of complicated expansion conditions, and the
6440 string was silently truncated, leading to an expansion error. It turns out
6441 that it is easier to change to an unlimited length (owing to other changes
6442 that have happened since this code was originally written) than to build
6443 structure for giving a limitation error. The length of the item that
6444 expands into the list of hosts is now unlimited.
6446 47. The lsearch lookup could not handle data where the length of text line was
6447 more than 4095 characters. Such lines were truncated, leading to shortened
6448 data being returned. It should now handle lines of any length.
6450 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
6451 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
6454 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
6455 replacing strings such as BIN_DIRECTORY in a source file: the replacement
6456 no longer happens in comment lines. A list of replacements is now placed
6457 at the head of all of the source files, except those whose only change is
6458 to replace PERL_COMMAND in the very first #! line.
6460 50. Replaced the slow insertion sort in queue.c, for sorting the list of
6461 messages on the queue, with a bottom-up merge sort, using code contributed
6462 by Michael Haardt. This should make operations like -bp somewhat faster on
6463 large queues. It won't affect queue runners, except when queue_run_in_order
6466 51. Installed eximstats 1.31 in the distribution.
6468 52. Added support for SRV lookups to the dnslookup router.
6470 53. If an ACL referred to $message_body or $message_body_end, the value was not
6471 reset for any messages that followed in the same SMTP session.
6473 54. The store-handling optimization for building very long strings was not
6474 differentiating between the different store pools. I don't think this
6475 actually made any difference in practice, but I've tidied it.
6477 55. While running the routers to verify a sender address, $sender_address
6478 was still set to the sender address. This is wrong, because when routing to
6479 send a bounce to the sender, it would be empty. Therefore, I have changed
6480 it so that, while verifying a sender address, $sender_address is set to <>.
6481 (There is no change to what happens when verifying a recipient address.)
6483 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
6484 target A or AAAA records (if not already returned) without resetting the
6485 qualify_single or search_parents options of the DNS resolver. These are
6486 inappropriate in this case because the targets of MX and SRV records must
6487 be FQDNs. A broken DNS record could cause trouble if it happened to have a
6488 target that, when qualified, matched something in the local domain. These
6489 two options are now turned off when doing these lookups.
6491 57. It seems that at least some releases of Reiserfs (which does not have the
6492 concept of a fixed number of inodes) returns zero and not -1 for the
6493 number of available inodes. This interacted badly with check_spool_inodes,
6494 which assumed that -1 was the "no such thing" setting. What I have done is
6495 to check that the total number of inodes is greater than zero before doing
6496 the test of how many are available.
6498 58. When a "warn" ACL statement has a log_message modifier, the message is
6499 remembered, and not repeated. This is to avoid a lot of repetition when a
6500 message has many recipients that cause the same warning to be written.
6501 However, Exim was preserving the list of already written lines for an
6502 entire SMTP session, which doesn't seem right. The memory is now reset if a
6503 new message is started.
6505 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6506 debugging output unless log_rewrite was also set.
6508 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6509 of (int)(handle) when we know that handle contains (void *)(-1).
6511 61. The Exim daemon panic-logs an error return when it closes the incoming
6512 connection. However "connection reset by peer" seems to be common, and
6513 isn't really an error worthy of noting specially, so that particular error
6516 62. When Exim is trying to find all the local interfaces, it used to panic and
6517 die if the ioctl to get the interface flags failed. However, it seems that
6518 on at least one OS (Solaris 9) it is possible to have an interface that is
6519 included in the list of interfaces, but for which you get a failure error
6520 for this call. This happens when the interface is not "plumbed" into a
6521 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6522 failure of the "get flags" call assumes that the interface is down.
6524 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6525 makes life easier for people who are doing arithmetic on fields extracted
6526 from dates, where you often get leading zeros that should not be
6527 interpreted as octal.
6529 64. Added qualify_domain to the redirect router, to override the global
6532 65. If a pathologically long header line contained very many addresses (the
6533 report of this problem mentioned 10 000) and each of them was rewritten,
6534 Exim could use up a very large amount of memory. (It kept on making new
6535 copies of the header line as it rewrote, and never released the old ones.)
6536 At the expense of a bit more processing, the header rewriting function has
6537 been changed so that it no longer eats memory in this way.
6539 66. The generation of the Received: header has been moved from the time that a
6540 message starts to be received, to the time that it finishes. The timestamp
6541 in the Received: header should now be very close to that of the <= log
6542 line. There are two side-effects of this change:
6544 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6545 logged header lines no longer include the local Received: line, because
6546 it has not yet been created. The same applies to a copy of the message
6547 that is returned to a non-SMTP sender when a message is rejected.
6549 (b) When a filter file is tested using -bf, no additional Received: header
6550 is added to the test message. After some thought, I decided that this
6553 This change does not affect the value of $received_for. It is still set
6554 after address rewriting, but before local_scan() is called.
6556 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6558 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6559 gave an unhelpful panic error message, and a defer error. I have managed to
6560 change this behaviour so that it now rejects any supplied certificate,
6561 which seems right, as the list of acceptable certificates is empty.
6563 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6564 gave an unhelpful defer error. I have not managed to make this reject any
6565 supplied certificates, but the error message it gives is "no certificate
6566 supplied", which is not helpful.
6568 70. exigrep's output now also includes lines that are not associated with any
6569 message, but which match the given pattern. Implemented by a patch from
6570 Martin Sluka, which also tidied up the Perl a bit.
6572 71. Recipient callout verification, like sender verification, was using <> in
6573 the MAIL FROM command. This isn't really the right thing, since the actual
6574 sender may affect whether the remote host accepts the recipient or not. I
6575 have changed it to use the actual sender in the callout; this means that
6576 the cache record is now keyed on a recipient/sender pair, not just the
6577 recipient address. There doesn't seem to be a real danger of callout loops,
6578 since a callout by the remote host to check the sender would use <>.
6579 [SEE ABOVE: changed after hitting problems.]
6581 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6582 temporary errors. However, in the case of such a code being given after
6583 the end of a data transmission (i.e. after ".") Exim was failing to write
6584 a retry record for the message. (Yes, there was some broken host that was
6585 actually sending 8xx at this point.)
6587 73. An unknown lookup type in a host list could cause Exim to panic-die when
6588 the list was checked. (An example that provoked this was putting <; in the
6589 middle of a list instead of at the start.) If this happened during a DATA
6590 ACL check, a -D file could be left lying around. This kind of configuration
6591 error no longer causes Exim to die; instead it causes a defer error. The
6592 incident is still logged to the main and panic logs.
6594 74. Buglet left over from Exim 3 conversion. The message "too many messages
6595 in one connection" was written to the rejectlog but not the mainlog, except
6596 when address rewriting (yes!) was being logged.
6598 75. Added write_rejectlog option.
6600 76. When a system filter was run not as root (that is, when system_filter_user
6601 was set), the values of the $n variables were not being returned to the
6602 main process; thus, they were not subsequently available in the $sn
6605 77. Added +return_path_on_delivery log selector.
6607 78. A connection timeout was being treated differently from recipients deferred
6608 when testing hosts_max_try with a message that was older than the host's
6609 retry timeout. (The host should not be counted, thus allowing all hosts to
6610 be tried at least once before bouncing.) This may have been the cause of an
6611 occasionally reported bug whereby a message would remain on the queue
6612 longer than the retry timeout, but would be bounced if a delivery was
6613 forced. I say "may" because I never totally pinned down the problem;
6614 setting up timeout/retry tests is difficult. See also the next item.
6616 79. The ultimate address timeout was not being applied to errors that involved
6617 a combination of host plus message (for example, a timeout on a MAIL
6618 command). When an address resolved to a number of possible hosts, and they
6619 were not all tried for each delivery (e.g. because of hosts_max_try), a
6620 message could remain on the queue longer than the retry timeout.
6622 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6625 81. Fixed an obscure SMTP outgoing bug which required at least the following
6626 conditions: (a) there was another message waiting for the same server;
6627 (b) the server returned 5xx to all RCPT commands in the first message so
6628 that the message was not completed; (c) the server dropped the connection
6629 or gave a negative response to the RSET that Exim sends to abort the
6630 transaction. The observed case was a dropped connection after DATA that had
6631 been sent in pipelining mode. That is, the server had advertised PIPELINING
6632 but was not implementing it correctly. The effect of the bug was incorrect
6633 behaviour, such as trying another host, and this could lead to a crash.
6639 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6640 and daemon.c were passed as pointers to ints; they should have been
6641 pointers to socklen_t variables (which are typically unsigned ints).
6643 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6646 3. Fixed a really odd bug that affected only the testing scheme; patching a
6647 certain fixed string in the binary changed the value of another string that
6648 happened to be identical to the end of the original first string.
6650 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6651 name", it returns that address as the IP address. On some operating
6652 systems (e.g. Solaris), it also passes back the IP address string as the
6653 "host name". However, on others (e.g. Linux), it passes back an empty
6654 string. Exim wasn't checking for this, and was changing the host name to an
6655 empty string, assuming it had been canonicalized.
6657 5. Although rare, it is permitted to have more than one PTR record for a given
6658 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6659 all the names associated with an address, because they do in Solaris.
6660 However, it seems that they do not in Linux for data that comes from the
6661 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6662 I found this out when I moved to a new Linux workstation and tried to run
6663 the Exim test suite.
6665 To get round this problem I have changed the code so that it now does its
6666 own call to the DNS to look up PTR records when searching for a host name.
6667 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6668 addresses that are only in /etc/hosts are still found.
6670 This behaviour is, however, controlled by an option called host_lookup_
6671 order, which defaults to "bydns:byaddr". If people want to use the other
6672 order, or indeed, just use one or the other means of lookup, they can
6673 specify it in this variable.
6675 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6676 some operating systems, this comes back from gethostbyaddr() as an empty
6677 string, and this is what Exim used to test for. However, it seems that in
6678 other systems, "." is yielded. Exim now tests for this case too.
6680 7. The values of check_spool_space and check_log_space are now held internally
6681 as a number of kilobytes instead of an absolute number of bytes. If a
6682 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6683 kilobyte. This means that much larger values can be stored.
6685 8. Exim monitor: an attempt to get the action menu when not actually pointing
6686 at a message produces an empty menu entitled "No message selected". This
6687 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6688 no entries in it ("Shell widget menu has zero width and/or height"). So I
6689 have added a single, blank menu entry in this case.
6691 9. Added ${quote_local_part.
6693 10. MIME decoding is now applied to the contents of Subject: header lines when
6696 11. Now that a reference to $sender_host_address automatically causes a reverse
6697 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6698 host lookup before query-style lookups in lists that might use this
6699 variable. This has therefore been abolished, and the "net-" prefix is no
6700 longer necessary for query-style lookups.
6702 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6703 have been a typo for LFLAGS, so it has been changed.
6705 13. The install script calls Exim with "-C /dev/null" in order to find the
6706 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6707 to be output. However, since Exim outputs its version number before the
6708 error, it didn't break the script. It just looked ugly. I fixed this by
6709 always allowing "-C /dev/null" if the caller is root.
6711 14. Ignore overlarge ACL variable number when reading spool file - insurance
6712 against a later release with more variables having written the file.
6714 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6715 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6716 this, as well as the form without the "IPv6" on the front.
6718 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6719 OS/Makefile-Darwin file.
6721 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6722 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6724 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6725 variable. However, some OS use ints for inodes. I've added cast to long int
6726 to get rid of the compiler warning.
6728 19. I had forgotten to lock out "/../" in configuration file names when
6729 ALT_CONFIG_PREFIX was set.
6731 20. Routers used for verification do not need to specify transports. However,
6732 if such a router generated a host list, and callout was configured, Exim
6733 crashed, because it could not find a port number from the (non-existent)
6734 transport. It now assumes port 25 in this circumstance.
6736 21. Added the -t option to exigrep.
6738 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6739 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6740 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6741 them is set, LOOKUP_LSEARCH is forced.
6743 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6745 24. Added sender and host information to the "rejected by local_scan()" log
6746 line; previously there was no indication of these.
6748 25. Added .include_if_exists.
6750 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6751 Linux. It turns out that not all file systems support this. Apparently some
6752 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6753 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6754 not supported on the file descriptor, is now ignored when Exim is trying to
6755 sync a directory. This applies only to Linux.
6757 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6759 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6760 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6763 29. Added support for maildirsize files from supplied patch (modified a bit).
6765 30. The use of :fail: followed by an empty string could lead Exim to respond to
6766 sender verification failures with (e.g.):
6768 550 Verification failed for <xxx>
6769 550 Sender verify failed
6771 where the first response line was missing the '-' that indicates it is not
6772 the final line of the response.
6774 31. The loop for finding the name of the user that called Exim had a hardwired
6775 limit of 10; it now uses the value of finduser_retries, which is used for
6776 all other user lookups.
6778 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6781 33. Exim was neglecting to zero errno before one call of strtol() when
6782 expanding a string and expecting an integer value. On some systems this
6783 resulted in spurious "integer overflow" errors. Also, it was casting the
6784 result into an int without checking.
6786 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6787 did not work. The code looks as if it has *never* worked, though it appears
6788 to have been documented since at least release 1.62. I have made it work.
6790 35. The "timeout_DNS" error in retry rules, also documented since at least
6791 1.62, also never worked. As it isn't clear exactly what this means, and
6792 clearly it isn't a major issue, I have abolished the feature by treating it
6793 as "timeout", and writing a warning to the main and panic logs.
6795 36. The display of retry rules for -brt wasn't always showing the error code
6798 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6799 timeout_connect_A, timeout_connect_MX.
6801 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6802 to the empty sender.
6804 39. The daemon was not analysing the content of -oX till after it had closed
6805 stderr and disconnected from the controlling terminal. This meant that any
6806 syntax errors were only noted on the panic log, and the return code from
6807 the command was 0. By re-arranging the code a little, I've made the
6808 decoding happen first, so such errors now appear on stderr, and the return
6809 code is 1. However, the actual setting up of the sockets still happens in
6810 the disconnected process, so errors there are still only recorded on the
6813 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6814 connections (as happens on some IP stacks) was logged at start up time as
6815 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6816 it from "IPv6 and IPv4", which means that two separate sockets are being
6819 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6820 says whether AF_INET or AF_INET6 was passed as an argument.
6822 42. Exiwhat output was messed up when time zones were included in log
6825 43. Exiwhat now gives more information about the daemon's listening ports,
6826 and whether -tls-on-connect was used.
6828 44. The "port" option of the smtp transport is now expanded.
6830 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6831 silently ignored. Now an error message is written to the main and panic
6834 46. There's a new ACL modifier called "logwrite" which writes to a log file
6835 as soon as it is encountered.
6837 47. Added $local_user_uid and $local_user_gid at routing time.
6839 48. Exim crashed when trying to verify a sender address that was being
6842 49. Exim was recognizing only a space character after ".include". It now also
6843 recognizes a tab character.
6845 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6846 extracting the relevant information from the specification. The man page no
6847 longer contains scrambled data for the -d option, and I've added a section
6848 at the front about calling Exim under different names.
6850 51. Added "extra_headers" argument to the "mail" command in filter files.
6852 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6855 53. Installed eximstats 1.29.
6857 54. Added transport_filter_timeout as a generic transport option.
6859 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6860 Cc: header lines. This was required by RFC 822, but it not required by RFC
6863 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6864 incoming messages that did not have them. Now it does so only if the
6865 message originates locally, that is, if there is no associated remote host
6866 address. When Resent- header lines are present, this applies to the Resent-
6867 lines rather than the non-Resent- lines.
6869 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6870 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6872 58. Messages for configuration errors now include the name of the main
6873 configuration file - useful now that there may be more than one file in a
6874 list (.included file names were always shown).
6876 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6877 for those rare installations that do not start the daemon as root or run it
6878 setuid root. I've cut out the call to initgroups() if the daemon is not
6881 60. The Exim user and group can now be bound into the binary as text strings
6882 that are looked up at the start of Exim's processing.
6884 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6886 62. Added $mailstore_basename variable.
6888 63. Installed patch to sieve.c from Michael Haardt.
6890 64. When Exim failed to open the panic log after failing to open the main log,
6891 the original message it was trying to log was written to stderr and debug
6892 output, but if they were not available (the usual case in production), it
6893 was lost. Now it is written to syslog before the two lines that record the
6894 failures to open the logs.
6896 65. Users' Exim filters run in subprocesses under the user's uid. It is
6897 possible for a "deliver" command or an alias in a "personal" command to
6898 provoke an address rewrite. If logging of address rewriting is configured,
6899 this fails because the process is not running as root or exim. There may be
6900 a better way of dealing with this, but for the moment (because 4.30 needs
6901 to be released), I have disabled address rewrite logging when running a
6902 filter in a non-root, non-exim process.
6908 1. The buildconfig auxiliary program wasn't quoting the value set for
6909 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6910 not defined. This bug was masked in 4.22 by the effect that was fixed in
6913 2. Some messages that were rejected after a message id was allocated were
6914 shown as "incomplete" by exigrep. It no longer does this for messages that
6915 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6917 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6918 have allow_domain_literals set, the ID did not get logged in the <= line.
6919 Domain literals are now always recognized in Message-ID: header lines.
6921 4. The first argument for a ${extract expansion item is the key name or field
6922 number. Leading and trailing spaces in this item were not being ignored,
6923 causing some misleading effects.
6925 5. When deliver_drop_privilege was set, single queue runner processes started
6926 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6927 same command in the process it spins off) were not dropping privilege.
6929 6. When the daemon running as "exim" started a queue runner, it always
6930 re-executed Exim in the spun-off process. This is a waste of effort when
6931 deliver_drop_privilege is set. The new process now just calls the
6932 queue-runner function directly.
6938 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6941 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6942 ignored. Though the use of -oP was forcing the writing of a pid file, it
6943 was always written to the default place.
6945 3. If the message "no IP address found for host xxxx" is generated during
6946 incoming verification, it is now followed by identification of the incoming
6947 connection (so you can more easily find what provoked it).
6949 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6951 5. Added some features to "harden" Exim a bit more against certain attacks:
6953 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6954 be put in Local/Makefile. This is like the never_users runtime option,
6955 but it cannot be overridden. The default setting is "root".
6957 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6958 prefix string with which any file named in a -C command line option
6961 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6962 is retained for -C and -D only if the caller of Exim is root. Without
6963 it, the exim user may also use -C and -D and retain privilege.
6965 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6966 command line option is disabled.
6968 6. Macro names set by the -D option must start with an upper case letter, just
6969 like macro names defined in the configuration file.
6971 7. Added "dereference=" facility to LDAP.
6973 8. Two instances of the typo "uknown" in the source files are fixed.
6975 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6976 the Configure-Makefile script screwed up while processing it.
6978 10. Incorporated PCRE 4.4.
6980 11. The SMTP synchronization check was not operating right at the start of an
6981 SMTP session. For example, it could not catch a HELO sent before the client
6982 waited for the greeting. There is now a check for outstanding input at the
6983 point when the greeting is written. Because of the duplex, asynchronous
6984 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6985 way, but not yet received, when the check is performed.
6987 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6988 on TCP/IP sockets, because this apparently causes some broken clients to
6991 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6992 unchanged) from the Cygwin maintainer.
6994 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6995 was supported instead of testing for mbx. Effectively a typo.
6997 15. The spa authenticator server code was not checking that the input it
6998 received was valid base64.
7000 16. The debug output line for the "set" modifier in ACLs was not showing the
7001 name of the variable that was being set.
7003 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
7005 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
7006 lookup on its own. Something else was needed to trigger the lookup. For
7007 example, a match in host_lookup or the need for a host name in a host list.
7008 Now, if $sender_host_name is referenced and the host name has not yet been
7009 looked up, a lookup is performed. If the lookup fails, the variable remains
7010 empty, and $host_lookup_failed is set to "1".
7012 19. Added "eqi" as a case-independent comparison operator.
7014 20. The saslauthd authentication condition could segfault if neither service
7015 nor realm was specified.
7017 21. If an overflowing value such as "2048M" was set for message_size_limit, the
7018 error message that was logged was misleading, and incoming SMTP
7019 connections were dropped. The message is now more accurate, and temporary
7020 errors are given to SMTP connections.
7022 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
7023 (except RSET) with a 421 error, until QUIT is received. However, it was
7024 failing to send a response to QUIT.
7026 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
7027 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
7028 ACL is now run after the helo_try_verify_hosts code.
7030 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
7031 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
7032 case-independent, so other case variants are also recognized). Apparently
7033 some systems use these upper case variants.
7035 25. If more than two messages were waiting for the same host, and a transport
7036 filter was specified for the transport, Exim sent two messages over the
7037 same TCP/IP connection, and then failed with "socket operation on non-
7038 socket" when it tried to send the third.
7040 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
7042 27. The extern definition of crypt16() in expand.c was not being excluded when
7043 the OS had its own crypt16() function.
7045 28. Added bounce_return_body as a new option, and bounce_return_size_limit
7046 as a preferred synonym for return_size_limit, both as an option and as an
7049 29. Added LIBS=-liconv to OS/Makefile-OSF1.
7051 30. Changed the default configuration ACL to relax the local part checking rule
7052 for addresses that are not in any local domains. For these addresses,
7053 slashes and pipe symbols are allowed within local parts, but the sequence
7054 /../ is explicitly forbidden.
7056 31. SPA server authentication was not clearing the challenge buffer before
7059 32. log_message in a "warn" ACL statement was writing to the reject log as
7060 well as to the main log, which contradicts the documentation and doesn't
7061 seem right (because no rejection is happening). So I have stopped it.
7063 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
7064 However, I am unable to do any testing of this.
7066 34. Fixed an infelicity in the appendfile transport. When checking directories
7067 for a mailbox, to see if any needed to be created, it was accidentally
7068 using path names with one or more superfluous leading slashes; tracing
7069 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
7071 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
7072 contents are added to the log line that is written for every discarded
7073 recipient. (Previously a log_message setting was ignored.)
7075 36. The ${quote: operator now quotes the string if it is empty.
7077 37. The install script runs exim in order to find its version number. If for
7078 some reason other than non-existence or emptiness, which it checks, it
7079 could not run './exim', it was installing it with an empty version number,
7080 i.e. as "exim-". This error state is now caught, and the installation is
7083 38. An argument was missing from the function that creates an error message
7084 when Exim fails to connect to the socket for saslauthd authentication.
7085 This could cause Exim to crash, or give a corrupted message.
7087 39. Added isip, isip4, and isip6 to ${if conditions.
7089 40. The ACL variables $acl_xx are now saved with the message, and can be
7090 accessed later in routers, transports, and filters.
7092 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
7093 strings in the file are not string-expanded.
7095 42. If a MAIL command specified a SIZE value that was too large to fit into an
7096 int variable, the check against message_size_limit failed. Such values are
7097 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
7098 day this will have to be increased, but I don't think I want to be around
7099 when emails are that large.
7106 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
7107 iconv() is not standard in FreeBSD.
7109 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
7110 IPv6 enabled. The observed symptom was a segmentation fault on return from
7111 the function os_common_find_running_interfaces() in src/os.c.
7113 3. In the check_special_case() function in daemon.c I had used "errno" as an
7114 argument name, which causes warnings on some systems. This was basically a
7115 typo, since it was named "eno" in the comments!
7117 4. The code that waits for the clock to tick (at a resolution of some fraction
7118 of a second) so as to ensure message-id uniqueness was always waiting for
7119 at least one whole tick, when it could have waited for less. [This is
7120 almost certainly not relevant at current processor speeds, where it is
7121 unlikely to ever wait at all. But we try to future-proof.]
7123 5. The function that sleeps for a time interval that includes fractions of a
7124 second contained a race. It did not block SIGALRM between setting the
7125 timer, and suspending (a couple of lines later). If the interval was short
7126 and the sigsuspend() was delayed until after it had expired, the suspension
7127 never ended. On busy systems this could lead to processes getting stuck for
7130 6. Some uncommon configurations may cause a lookup to happen in a queue runner
7131 process, before it forks any delivery processes. The open lookup caching
7132 mechanism meant that the open file or database connection was passed into
7133 the delivery process. The problem was that delivery processes always tidy
7134 up cached lookup data. This could cause a problem for the next delivery
7135 process started by the queue runner, because the external queue runner
7136 process does not know about the closure. So the next delivery process
7137 still has data in the lookup cache. In the case of a file lookup, there was
7138 no problem because closing a file descriptor in a subprocess doesn't affect
7139 the parent. However, if the lookup was caching a connection to a database,
7140 the connection was closed, and the second delivery process was likely to
7141 see errors such as "PGSQL: query failed: server closed the connection
7142 unexpectedly". The problem has been fixed by closing all cached lookups
7143 in a queue runner before running a delivery process.
7145 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
7146 seem to have the "const" qualifier which it has on other OS. I've
7149 8. Change 4.21/2 was too strict. It is only if there are two authenticators
7150 *of the same type* (client or server) with the same public name that an
7151 error should be diagnosed.
7153 9. When Exim looked up a host name for an IP address, but failed to find the
7154 original IP address when looking up the host name (a safety check), it
7155 output the message "<ip address> does not match any IP for NULL", which was
7156 confusing, to say the least. The bug was that the host name should have
7157 appeared instead of "NULL".
7159 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
7160 user that is built into the binary, and the -C or -D options is used, root
7161 privilege is dropped before the configuration file is read. In addition,
7162 logging is switched to stderr instead of the normal log files. If the
7163 configuration then re-defines the Exim user, the unprivileged environment
7164 is probably not what is expected, so Exim logs a panic warning message (but
7167 However, if deliver_drop_privilege is set, the unprivileged state may well
7168 be exactly what is intended, so the warning has been cut out in that case,
7169 and Exim is allowed to try to write to its normal log files.
7175 1. smtp_return_error_details was not giving details for temporary sender
7176 or receiver verification errors.
7178 2. Diagnose a configuration error if two authenticators have the same public
7181 3. Exim used not to create the message log file for a message until the first
7182 delivery attempt. This could be confusing when incoming messages were held
7183 for policy or load reasons. The message log file is now created at the time
7184 the message is received, and an initial "Received" line is written to it.
7186 4. The automatically generated man page for command line options had a minor
7187 bug that caused no ill effects; however, a more serious problem was that
7188 the procedure for building the man page automatically didn't always
7189 operate. Consequently, release 4.20 contains an out-of-date version. This
7190 shouldn't happen again.
7192 5. When building Exim with embedded Perl support, the script that builds the
7193 Makefile was calling 'perl' to find its compile-time parameters, ignoring
7194 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
7196 6. The freeze_tell option was not being used for messages that were frozen on
7197 arrival, either by an ACL or by local_scan().
7199 7. Added the smtp_incomplete_transaction log selector.
7201 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
7202 was accepting AUTH without a new EHLO.
7204 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
7205 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
7208 10. Logging of TCP/IP connections (when configured) now happens in the main
7209 daemon process instead of the child process, so that the TCP/IP connection
7210 count is more accurate (but it can never be perfect).
7212 11. The use of "drop" in a nested ACL was not being handled correctly in the
7213 outer ACL. Now, if condition failure induced by the nested "drop" causes
7214 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
7215 or "require"), the connection is dropped.
7217 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
7218 that yield "discard" can now be used with an "accept" or a "discard" verb,
7219 but an error is generated for any others (because I can't see a useful way
7220 to define what should happen).
7222 13. When an ACL is read dynamically from a file (or anywhere else), the lines
7223 are now processed in the same way as lines in the Exim configuration file.
7224 In particular, continuation lines are supported.
7226 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
7228 15. Added -ti meaning -t -i.
7230 16. Check for letters, digits, hyphens, and dots in the names of dnslist
7231 domains, and warn by logging if others are found.
7233 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
7234 returned from GIFCONF when Exim is trying to find the list of interfaces on
7235 a host. The code in os.c has been modified to copy each ifreq to an aligned
7236 structure in all cases.
7238 Also, in some cases, the returned ifreq's were being copied to a 'struct
7239 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
7240 means the last couple of bytes of an IPv6 address could be chopped if the
7241 ifreq contained only a normal sockaddr (14 bytes storage).
7243 18. Named domain lists were not supported in the hosts_treat_as_local option.
7244 An entry such as +xxxx was not recognized, and was treated as a literal
7247 19. Ensure that header lines added by a DATA ACL are included in the reject log
7248 if the ACL subsequently rejects the message.
7250 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
7251 MD5 (which is deprecated).
7253 21. When testing a filter file using -bf, Exim was writing a message when it
7254 took the sender from a "From " line in the message, but it was not doing so
7255 when it took $return_path from a Return-Path: header line. It now does.
7257 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
7258 with a valid header line field name (a series of printing characters
7259 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
7261 23. Changed "disc" in the source to "disk" to conform to the documentation and
7262 the book and for uniformity.
7264 24. Ignore Sendmail's -Ooption=value command line item.
7266 25. When execve() failed while trying to run a command in a pipe transport,
7267 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
7268 could be confused with a return value of 69 from the command itself. This
7269 has been changed to 127, the value the shell returns if it is asked to run
7270 a non-existent command. The wording for the related log line suggests a
7271 non-existent command as the problem.
7273 26. If received_header_text expands to an empty string, do not add a Received:
7274 header line to the message. (Well, it adds a token one on the spool, but
7275 marks it "old" so that it doesn't get used or transmitted.)
7277 27. Installed eximstats 1.28 (addition of -nt option).
7279 28. There was no check for failure on the call to getsockname() in the daemon
7280 code. This can fail if there is a shortage of resources on the system, with
7281 ENOMEM, for example. A temporary error is now given on failure.
7283 29. Contrary to the C standard, it seems that in some environments, the
7284 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
7285 program. Exim now does this explicitly; it affects the formatting of
7286 timestamps using strftime().
7288 30. If exiqsumm was given junk data, it threw up some uninitialized variable
7289 complaints. I've now initialized all the variables, to avoid this.
7291 32. Header lines added by a system filter were not being "seen" during
7292 transport-time rewrites.
7294 33. The info_callback() function passed to OpenSSL is set up with type void
7295 (*)(SSL *, int, int), as described somewhere. However, when calling the
7296 function (actually a macro) that sets it up, the type void(*)() is
7297 expected. I've put in a cast to prevent warnings from picky compilers.
7299 34. If a DNS black list lookup found a CNAME record, but there were no A
7300 records associated with the domain it pointed at, Exim crashed.
7302 35. If a DNS black list lookup returned more than one A record, Exim ignored
7303 all but the first. It now scans all returned addresses if a particular IP
7304 value is being sought. In this situation, the contents of the
7305 $dnslist_value variable are a list of all the addresses, separated by a
7308 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
7309 to accept a host name and all its aliases if the forward lookup for any of
7310 them yielded the IP address of the incoming connection. Now it accepts only
7311 those names whose forward lookup yields the correct IP address. Any other
7312 names are discarded. This closes a loophole whereby a rogue DNS
7313 administrator could create reverse DNS records to break through a
7314 wildcarded host restriction in an ACL.
7316 37. If a user filter or a system filter that ran in a subprocess used any of
7317 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
7318 the wrong values were passed to the pipe command ($thisaddress had the
7319 value of $0, $0 had the value of $1, etc). This bug was introduced by
7320 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
7322 38. Improved the line breaking for long SMTP error messages from ACLs.
7323 Previously, if there was no break point between 40 and 75 characters, Exim
7324 left the rest of the message alone. Two changes have been made: (a) I've
7325 reduced the minimum length to 35 characters; (b) if it can't find a break
7326 point between 35 and 75 characters, it looks ahead and uses the first one
7327 that it finds. This may give the occasional overlong line, but at least the
7328 remaining text gets split now.
7330 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
7331 file descriptors might be low, and that setting 1000 would always raise it.
7332 It turns out that in some environments, the limit is already over 1000 and
7333 that lowering it causes trouble. So now Exim takes care not to decrease it.
7335 40. When delivering a message, the value of $return_path is set to $sender_
7336 address at the start of routing (routers may change the value). By an
7337 oversight, this default was not being set up when an address was tested by
7338 -bt or -bv, which affected the outcome if any router or filter referred to
7341 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
7342 log only when "message" or "log_message" are set. However, if one of the
7343 conditions was an address verification, or a call to a nested ACL, the
7344 messages generated by the underlying test were being passed through. This
7345 no longer happens. The underlying message is available in $acl_verify_
7346 message for both "message" and "log_message" expansions, so it can be
7347 passed through if needed.
7349 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
7350 new expansion $bh_ to give the encoded byte string without charset
7351 translation. Translation happens only if iconv() is available; HAVE_ICONV
7352 indicates this at build time. HEADERS_CHARSET gives the charset to
7353 translate to; headers_charset can change it in the configuration, and
7354 "headers charset" can change it in an individual filter file.
7356 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
7357 that creates RFC 2047 encoded "words" labels them as that charset instead
7358 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
7359 expansion operator; (ii) when Exim creates a From: line for a local
7360 message; (iii) when a header line is rewritten to include a "phrase" part.
7362 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
7363 buggy, causing it to skip the first lines of messages whose message ID
7364 ended in 'D'. This would not have bitten before Exim release 4.14, because
7365 message IDs were unlikely to end in 'D' before then. The effect was to have
7366 incorrect size information for certain domains.
7368 45. #include "config.h" was missing at the start of the crypt16.c module. This
7369 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
7372 46. If there was a timeout during a "random" callout check, Exim treated it as
7373 a failure of the random address, and carried on sending RSET and the real
7374 address. If the delay was just some slowness somewhere, the response to the
7375 original RCPT would be taken as a response to RSET and so on, causing
7376 mayhem of various kinds.
7378 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
7379 when I implemented it. It didn't allow for the fact that some option values
7380 may legitimately be negative (e.g. size_addition), and it didn't even do
7381 the right test for positive values.
7383 48. Domain names in DNS records are case-independent. Exim always looks them up
7384 in lower case. Some resolvers return domain names in exactly the case they
7385 appear in the zone file, that is, they may contain uppercase letters. Not
7386 all resolvers do this - some return always lower case. Exim was treating a
7387 change of case by a resolver as a change of domain, similar to a widening
7388 of a domain abbreviation. This triggered its re-routing code and so it was
7389 trying to route what was effectively the same domain again. This normally
7390 caused routing to fail (because the router wouldn't handle the domain
7391 twice). Now Exim checks for this case specially, and just changes the
7392 casing of the domain that it ultimately uses when it transmits the message
7395 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
7398 50. If a filter generated a file delivery with a non-absolute name (possible if
7399 no home directory exists for the router), the forbid_file option was not
7402 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
7403 the existing equality matching.
7405 52. Exim was using ints instead of ino_t variables in some places where it was
7406 dealing with inode numbers.
7408 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
7409 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
7410 called TMPDIR, and if it finds it is different, it changes its value.
7412 54. The smtp_printf() function is now made available to local_scan() so
7413 additional output lines can be written before returning. There is also an
7414 smtp_fflush() function to enable the detection of a dropped connection.
7415 The variables smtp_input and smtp_batched_input are exported to
7418 55. Changed the default runtime configuration: the message "Unknown user"
7419 has been removed from the ACL, and instead placed on the localuser router,
7420 using the cannot_route_message feature. This means that any verification
7421 failures that generate their own messages won't get overridden. Similarly,
7422 the "Unrouteable address" message that was in the ACL for unverifiable
7423 relay addresses has also been removed.
7425 56. Added hosts_avoid_esmtp to the smtp transport.
7427 57. The exicyclog script was not checking for the esoteric option
7428 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
7429 will work only if exicyclog is run under the appropriate euid.
7431 58. Following a discussion on the list, the rules by which Exim recognises line
7432 endings on incoming messages have been changed. The -dropcr and drop_cr
7433 options are now no-ops, retained only for backwards compatibility. The
7434 following line terminators are recognized: LF CRLF CR. However, special
7435 processing applies to CR:
7437 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
7438 nor a local message in the state where . is a terminator.
7440 (ii) If a bare CR is encountered in a header line, an extra space is added
7441 after the line terminator so as not to end the header. The reasoning
7442 behind this is that bare CRs in header lines are most likely either
7443 to be mistakes, or people trying to play silly games.
7445 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
7446 was being incorrectly given as 18 bytes larger than it should have been.
7447 This is a VOB (very old bug).
7449 60. This may never have affected anything current, but just in case it has:
7450 When the local host is found other than at the start of a list of hosts,
7451 the local host, those with the same MX, and any that follow, are discarded.
7452 When the list in question was part of a longer list of hosts, the following
7453 hosts (not currently being processed) were also being discarded. This no
7454 longer happens. I'm not sure if this situation could ever has previously
7457 61. Added the "/MX" feature to lists of hosts in the manualroute and query
7460 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
7461 header. This is something that is recommended in a new Internet Draft, and
7462 is something that is documented as being done by Sendmail. There are two
7463 possible values. For messages generated by the autoreply transport, Exim
7466 Auto-Submitted: auto-replied
7468 whereas for all other generated messages (e.g. bounces) it adds
7470 Auto-Submitted: auto-generated
7472 63. The "personal" condition in filters now includes a test for the
7473 Auto-Submitted: header. If it contains the string "auto-" the message it
7474 not considered personal.
7476 64. Added rcpt_include_affixes as a generic transport option.
7478 65. Added queue_only_override (default true).
7480 66. Added the syslog_duplication option.
7482 67. If what should have been the first header line of a message consisted of
7483 a space followed by a colon, Exim was mis-interpreting it as a header line.
7484 It isn't of course - it is syntactically invalid and should therefore be
7485 treated as the start of the message body. The misbehaviour could have
7486 caused a number of strange effects, including loss of data in subsequent
7487 header lines, and spool format errors.
7489 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
7490 client host had authenticated. This control can now be exercised by an ACL
7491 for more flexibility.
7493 69. By default, callouts do not happen when testing with -bh. There is now a
7494 variant, -bhc, which does actually run the callout code, including
7495 consulting and updating the callout cache.
7497 70. Added support for saslauthd authentication, courtesy of Alexander
7500 71. If statvfs() failed on the spool or log directories while checking their
7501 size for availability, Exim confusingly gave the error "space shortage".
7502 Furthermore, in debugging mode it crashed with a floating point exception.
7503 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7504 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7505 serious problem, Exim now writes to the main and panic logs when this
7506 happens, with details of the failure. It then refuses to accept the
7507 incoming message, giving the message "spool directory problem" or "log
7508 directory problem" with a 421 code for SMTP messages.
7510 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7511 0, 1, and 2 exist, because some OS complain for execs without them (see
7512 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7513 descriptors. However, the code omitted to check that the open succeeded,
7514 causing mysterious errors if for some reason the permissions on /dev/null
7515 got screwed. Now Exim writes a message to the main and panic logs, and
7516 bombs out if it can't open /dev/null.
7518 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7519 interact so that it is all more flexible. It is supposed to remain
7520 backwards compatible. Also added extra_local_interfaces.
7522 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7523 to bomb out with an assertion failure - to the client this appears as a
7524 connection drop. This problem occurs in the part of the code that was taken
7525 from the Samba project. Fortunately, the assertion is in a very simple
7526 function, so I have fixed this by reproducing the function inline in the
7527 one place where it is called, and arranging for authentication to fail
7528 instead of killing the process with assert().
7530 75. The SPA client code was not working when the server requested OEM rather
7531 than Unicode encoding.
7533 76. Added code to make require_files with a specific uid setting more usable in
7534 the case where statting the file as root fails - usually a non-root-mounted
7535 NFS file system. When this happens and the failure is EACCES, Exim now
7536 forks a subprocess and does the per-uid checking as the relevant uid.
7538 77. Added process_log_path.
7540 78. If log_file_path was not explicitly set, a setting of check_log_space or
7541 check_log_inodes was ignored.
7543 79. If a space check for the spool or log partitions fails, the incident is now
7544 logged. Of course, in the latter case the data may get lost...
7546 80. Added the %p formatting code to string_format() so that it can be used to
7547 print addresses in debug_print(). Adjusted all the address printing in the
7548 debugging in store.c to use %p rather than %d.
7550 81. There was a concern that a line of code in smtp_in.c could overflow a
7551 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7552 initially expressed, the concern was not well-founded, because trailing
7553 spaces are removed early. However, if the trailing spaces were followed by
7554 a NULL, they did not get removed, so the overflow was possible. Two fixes
7557 (a) I re-wrote the offending code in a cleaner fashion.
7558 (b) If an incoming SMTP command contains a NULL character, it is rejected
7561 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7562 runs initgroups(), so that if the Exim user is in any additional groups,
7563 they will be used during message reception.
7569 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob