1 Change log file for Exim from version 4.21
2 ------------------------------------------
3 This document describes *changes* to previous versions, that might
4 affect Exim's operation, with an unchanged configuration file. For new
5 options, and new features, see the NewStuff file next to this ChangeLog.
11 JH/01 OpenSSL: With debug enabled output keying information sufficient, server
12 side, to decode a TLS 1.3 packet capture.
14 JH/02 OpenSSL: Suppress the sending of (stateful) TLS1.3 session tickets.
15 Previously the default library behaviour applied, sending two, each in
18 JH/03 Debug output for ACL now gives the config file name and line number for
21 JH/04 The default received_header_text now uses the RFC 8314 tls cipher clause.
23 JH/05 DKIM: ensure that dkim_domain elements are lowercased before use.
25 JH/06 Fix buggy handling of autoreply bounce_return_size_limit, and a possible
26 buffer overrun for (non-chunking) other transports.
28 JH/07 GnuTLS: Our use of late (post-handshake) certificate verification, under
29 TLS1.3, means that a server rejecting a client certificate is not visible
30 to the client until the first read of encrypted data (typically the
31 response to EHLO). Add detection for that case and treat it as a failed
32 TLS connection attempt, so that the normal retry-in-clear can work (if
35 JB/01 Bug 2375: fix expansions of 822 addresses having comments in local-part
36 and/or domain. Found and fixed by Jason Betts.
38 JH/08 Add hardening against SRV & TLSA lookups the hit CNAMEs (a nonvalid
39 configuration). If a CNAME target was not a wellformed name pattern, a
42 JH/09 Logging: Fix initial listening-on line for multiple ports for an IP when
43 the OS reports them interleaved with other addresses.
45 JH/10 OpenSSL: Fix aggregation of messages. Previously, when PIPELINING was
46 used both for input and for a verify callout, both encrypted, SMTP
47 responses being sent by the server could be lost. This resulted in
48 dropped connections and sometimes bounces generated by a peer sending
51 JH/11 Harden plaintext authenticator against a badly misconfigured client-send
52 string. Previously it was possible to cause undefined behaviour in a
53 library routine (usually a crash). Found by "zerons".
55 JH/12 Bug 2384: fix "-bP smtp_receive_timeout". Previously it returned no
58 JH/13 Bug 2386: Fix builds with Dane under LibreSSL 2.9.0 onward. Some old
59 API was removed, so update to use the newer ones.
61 JH/14 Bug 1891: Close the log file if receiving a non-smtp message, without
62 any timeout set, is taking a long time. Previously we would hang on to a
63 rotated logfile "forever" if the input was arriving with long gaps
64 (a previous attempt to fix addressed lack, for a long time, of initial
67 HS/01 Bug 2390: Use message_id for tempfile creation to avoid races in a
68 shared (NFS) environment. The length of the tempfile name is now
69 4 + 16 ("hdr.$message_exim_id") which might break on file
70 systems which restrict the file name length to lower values.
73 HS/02 Bug 2390: Use message_id for tempfile creation to avoid races in a
74 shared (NFS) environment.
76 HS/03 Bug 2392: exigrep does case sensitive *option* processing (as it
77 did for all versions <4.90). Notably -M, -m, --invert, -I may be
80 JH/15 Use unsigned when creating bitmasks in macros, to avoid build errors
81 on some platforms for bit 31.
83 JH/16 GnuTLS: rework ciphersuite strings under recent library versions. Thanks
84 to changes apparently associated with TLS1.3 handling some of the APIs
85 previously used were either nonfunctional or inappropriate. Strings
86 like TLS1.3:ECDHE_SECP256R1__RSA_PSS_RSAE_SHA256__AES_256_GCM__AEAD:256
87 and TLS1.2:ECDHE_SECP256R1__RSA_SHA256__AES_128_CBC__SHA256:128 replace
88 the previous TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256 .
89 This affects log line X= elements, the $tls_{in,out}_cipher variables,
90 and the use of specific cipher names in the encrypted= ACL condition.
92 JH/17 OpenSSL: the default openssl_options now disables ssl_v3.
94 JH/18 GnuTLS: fix $tls_out_ocsp under hosts_request_ocsp. Previously the
95 verification result was not updated unless hosts_require_ocsp applied.
97 JH/19 Bug 2398: fix listing of a named-queue. Previously, even with the option
98 queue_list_requires_admin set to false, non-admin users were denied the
101 JH/20 Bug 2389: fix server advertising of usable certificates, under GnuTLS in
102 directory-of-certs mode. Previously they were advertised despite the
105 JH/21 The smtp transport option "hosts_noproxy_tls" is now unset by default.
106 A single TCP connection by a client will now hold a TLS connection open
107 for multiple message deliveries, by default. Previoud the default was to
110 JH/22 The smtp transport option "hosts_try_dane" now enables all hosts by
111 default. If built with the facility, DANE will be used. The facility
112 SUPPORT_DANE is now enabled in the prototype build Makefile "EDITME".
114 JH/23 The build default is now for TLS to be included; the SUPPORT_TLS define
115 is replaced with DISABLE_TLS. Either USE_GNUTLS or (the new) USE_OPENSSL
116 must be defined and you must still, unless you define DISABLE_TLS, manage
117 the the include-dir and library-file requirements that go with that
118 choice. Non-TLS builds are still supported.
120 JH/24 Fix duplicated logging of peer name/address, on a transport connection-
123 JH/25 The smtp transport option "hosts_try_fastopen" now enables all hosts by
124 default. If the platform supports and has the facility enabled, it will
125 be requested on all coneections.
127 JH/26 The PIPE_CONNECT facility is promoted from experimental status and is now
128 controlled by the build-time option SUPPORT_PIPE_CONNECT.
130 PP/01 Unbreak heimdal_gssapi, broken in 4.92.
132 JH/27 Bug 2404: Use the main-section configuration option "dsn_from" for
133 success-DSN messages. Previously the From: header was always the default
134 one for these; the option was ignored.
136 JH/28 Fix the timeout on smtp response to apply to the whole response.
137 Previously it was reset for every read, so a teergrubing peer sending
138 single bytes within the time limit could extend the connection for a
139 long time. Credit to Qualsys Security Advisory Team for the discovery.
141 JH/29 Fix DSN Final-Recipient: field. Previously it was the post-routing
142 delivery address, which leaked information of the results of local
143 forwarding. Change to the original envelope recipient address, per
146 JH/30 Bug 2411: Fix DSN generation when RFC 3461 failure notification is
147 requested. Previously not bounce was generated and a log entry of
148 error ignored was made.
150 JH/31 Avoid re-expansion in ${sort } expansion. (CVE-2019-13917)
152 JH/32 Introduce a general tainting mechanism for values read from the input
153 channel, and values derived from them. Refuse to expand any tainted
154 values, to catch one form of exploit.
156 JH/33 Bug 2413: Fix dkim_strict option. Previously the expansion result
157 was unused and the unexpanded text used for the test. Found and
158 fixed by Ruben Jenster.
160 JH/34 Fix crash after TLS shutdown. When the TCP/SMTP channel was left open,
161 an attempt to use a TLS library read routine dereffed a nul pointer,
164 JH/35 Bug 2409: filter out-of-spec chars from callout response before using
165 them in our smtp response.
167 JH/36 Have the general router option retry_use_local_part default to true when
168 any of the restrictive preconditions are set (to anything). Previously it
169 was only for check_local user. The change removes one item of manual
170 configuration which is required for proper retries when a remote router
171 handles a subset of addresses for a domain.
173 JH/37 Appendfile: when evaluating quota use (non-quota_size_regex) take the file
174 link count into consideration.
176 HS/04 Fix handling of very log lines in -H files. If a -<key> <value> line
177 caused the extension of big_buffer, the following lines were ignored.
179 JH/38 Bug 1395: Teach the DNS negative-cache about TTL value from the SOA in
180 accordance with RFC 2308. Previously there was no expiry, so a longlived
181 receive process (eg. due to ACL delays) versus a short SOA value could
184 HS/05 Handle trailing backslash gracefully. (CVE-2019-15846)
186 JH/39 Promote DMARC support to mainline.
188 JH/40 Bug 2452: Add a References: header to DSNs.
190 JH/41 With GnuTLS 3.6.0 (and later) do not attempt to manage Diffie-Hellman
191 parameters. The relevant library call is documented as "Deprecated: This
192 function is unnecessary and discouraged on GnuTLS 3.6.0 or later. Since
193 3.6.0, DH parameters are negotiated following RFC7919."
199 JH/01 Remove code calling the customisable local_scan function, unless a new
200 definition "HAVE_LOCAL_SCAN=yes" is present in the Local/Makefile.
202 JH/02 Bug 1007: Avoid doing logging from signal-handlers, as that can result in
203 non-signal-safe functions being used.
205 JH/03 Bug 2269: When presented with a received message having a stupidly large
206 number of DKIM-Signature headers, disable DKIM verification to avoid
207 a resource-consumption attack. The limit is set at twenty.
209 JH/04 Add variables $arc_domains, $arc_oldest_pass for ARC verify. Fix the
210 report of oldest_pass in ${authres } in consequence, and separate out
211 some descriptions of reasons for verification fail.
213 JH/05 Bug 2273: Cutthrough delivery left a window where the received messsage
214 files in the spool were present and unlocked. A queue-runner could spot
215 them, resulting in a duplicate delivery. Fix that by doing the unlock
216 after the unlink. Investigation by Tim Stewart. Take the opportunity to
217 add more error-checking on spoolfile handling while that code is being
220 PP/01 Refuse to open a spool data file (*-D) if it's a symlink.
221 No known attacks, no CVE, this is defensive hardening.
223 JH/06 Bug 2275: The MIME ACL unlocked the received message files early, and
224 a queue-runner could start a delivery while other operations were ongoing.
225 Cutthrough delivery was a common victim, resulting in duplicate delivery.
226 Found and investigated by Tim Stewart. Fix by using the open message data
227 file handle rather than opening another, and not locally closing it (which
228 releases a lock) for that case, while creating the temporary .eml format
229 file for the MIME ACL. Also applies to "regex" and "spam" ACL conditions.
231 JH/07 Bug 177: Make a random-recipient callout success visible in ACL, by setting
232 $sender_verify_failure/$recipient_verify_failure to "random".
234 JH/08 When generating a selfsigned cert, use serial number 1 since zero is not
237 JH/09 Bug 2274: Fix logging of cmdline args when starting in an unlinked cwd.
238 Previously this would segfault.
240 JH/10 Fix ARC signing for case when DKIM signing failed. Previously this would
243 JH/11 Bug 2264: Exim now only follows CNAME chains one step by default. We'd
244 like zero, since the resolver should be doing this for us, But we need one
245 as a CNAME but no MX presence gets the CNAME returned; we need to check
246 that doesn't point to an MX to declare it "no MX returned" rather than
247 "error, loop". A new main option is added so the older capability of
248 following some limited number of chain links is maintained.
250 JH/12 Add client-ip info to non-pass iprev ${authres } lines.
252 JH/13 For receent Openssl versions (1.1 onward) use modern generic protocol
253 methods. These should support TLS 1.3; they arrived with TLS 1.3 and the
254 now-deprecated earlier definitions used only specified the range up to TLS
255 1.2 (in the older-version library docs).
257 JH/14 Bug 2284: Fix DKIM signing for body lines starting with a pair of dots.
259 JH/15 Rework TLS client-side context management. Stop using a global, and
260 explicitly pass a context around. This enables future use of TLS for
261 connections to service-daemons (eg. malware scanning) while a client smtp
262 connection is using TLS; with cutthrough connections this is quite likely.
264 JH/16 Fix ARC verification to do AS checks in reverse order.
266 JH/17 Support a "tls" option on the ${readsocket } expansion item.
268 JH/18 Bug 2287: Fix the protocol name (eg utf8esmtp) for multiple messages
269 using the SMTPUTF8 option on their MAIL FROM commands, in one connection.
270 Previously the "utf8" would be re-prepended for every additional message.
272 JH/19 Reject MAIL FROM commands with SMTPUTF8 when the facility was not advertised.
273 Previously thery were accepted, resulting in issues when attempting to
274 forward messages to a non-supporting MTA.
276 PP/02 Let -n work with printing macros too, not just options.
278 JH/20 Bug 2296: Fix cutthrough for >1 address redirection. Previously only
279 one parent address was copied, and bogus data was used at delivery-logging
280 time. Either a crash (after delivery) or bogus log data could result.
281 Discovery and analysis by Tim Stewart.
283 PP/03 Make ${utf8clean:} expansion operator detect incomplete final character.
284 Previously if the string ended mid-character, we did not insert the
285 promised '?' replacement.
287 PP/04 Documentation: current string operators work on bytes, not codepoints.
289 JH/21 Change as many as possible of the global flags into one-bit bitfields; these
290 should pack well giving a smaller memory footprint so better caching and
291 therefore performance. Group the declarations where this can't be done so
292 that the byte-sized flag variables are not interspersed among pointer
293 variables, giving a better chance of good packing by the compiler.
295 JH/22 Bug 1896: Fix the envelope from for DMARC forensic reports to be possibly
296 non-null, to avoid issues with sites running BATV. Previously reports were
297 sent with an empty envelope sender so looked like bounces.
299 JH/23 Bug 2318: Fix the noerror command within filters. It wasn't working.
300 The ignore_error flag wasn't being returned from the filter subprocess so
301 was not set for later routers. Investigation and fix by Matthias Kurz.
303 JH/24 Bug 2310: Raise a msg:fail:internal event for each undelivered recipient,
304 and a msg:complete for the whole, when a message is manually removed using
305 -Mrm. Developement by Matthias Kurz, hacked on by JH.
307 JH/25 Avoid fixed-size buffers for pathnames in DB access. This required using
308 a "Gnu special" function, asprintf() in the DB utility binary builds; I
309 hope that is portable enough.
311 JH/26 Bug 2311: Fix DANE-TA verification under GnuTLS. Previously it was also
312 requiring a known-CA anchor certificate; make it now rely entirely on the
313 TLSA as an anchor. Checking the name on the leaf cert against the name
314 on the A-record for the host is still done for TA (but not for EE mode).
316 JH/27 Fix logging of proxy address. Previously, a pointless "PRX=[]:0" would be
317 included in delivery lines for non-proxied connections, when compiled with
318 SUPPORT_SOCKS and running with proxy logging enabled.
320 JH/28 Bug 2314: Fire msg:fail:delivery event even when error is being ignored.
321 Developement by Matthias Kurz, tweaked by JH. While in that bit of code,
322 move the existing event to fire before the normal logging of message
323 failure so that custom logging is bracketed by normal logging.
325 JH/29 Bug 2322: A "fail" command in a non-system filter (file) now fires the
326 msg:fail:internal event. Developement by Matthias Kurz.
328 JH/30 Bug 2329: Increase buffer size used for dns lookup from 2k, which was
329 far too small for todays use of crypto signatures stored there. Go all
330 the way to the max DNS message size of 64kB, even though this might be
331 overmuch for IOT constrained device use.
333 JH/31 Fix a bad use of a copy function, which could be used to pointlessly
334 copy a string over itself. The library routine is documented as not
335 supporting overlapping copies, and on MacOS it actually raised a SIGABRT.
337 JH/32 For main options check_spool_space and check_inode_space, where the
338 platform supports 64b integers, support more than the previous 2^31 kB
339 (i.e. more than 2 TB). Accept E, P and T multipliers in addition to
340 the previous G, M, k.
342 JH/33 Bug 2338: Fix the cyrus-sasl authenticator to fill in the
343 $authenticated_fail_id variable on authentication failure. Previously
346 JH/34 Increase RSA keysize of autogen selfsign cert from 1024 to 2048. RHEL 8.0
347 OpenSSL didn't want to use such a weak key. Do for GnuTLS also, and for
348 more-modern GnuTLS move from GNUTLS_SEC_PARAM_LOW to
349 GNUTLS_SEC_PARAM_MEDIUM.
351 JH/35 OpenSSL: fail the handshake when SNI processing hits a problem, server
352 side. Previously we would continue as if no SNI had been received.
354 JH/36 Harden the handling of string-lists. When a list consisted of a sole
355 "<" character, which should be a list-separator specification, we walked
356 off past the nul-terimation.
358 JH/37 Bug 2341: Send "message delayed" warning MDNs (restricted to external
359 causes) even when the retry time is not yet met. Previously they were
360 not, meaning that when (say) an account was over-quota and temp-rejecting,
361 and multiple senders' messages were queued, only one sender would get
362 notified on each configured delay_warning cycle.
364 JH/38 Bug 2351: Log failures to extract envelope addresses from message headers.
366 JH/39 OpenSSL: clear the error stack after an SSL_accept(). With anon-auth
367 cipher-suites, an error can be left on the stack even for a succeeding
368 accept; this results in impossible error messages when a later operation
371 AM/01 Bug 2359: GnuTLS: repeat lowlevel read and write operations while they
372 return error codes indicating retry. Under TLS1.3 this becomes required.
374 JH/40 Fix the feature-cache refresh for EXPERIMENTAL_PIPE_CONNECT. Previously
375 it only wrote the new authenticators, resulting in a lack of tracking of
376 peer changes of ESMTP extensions until the next cache flush.
378 JH/41 Fix the loop reading a message header line to check for integer overflow,
379 and more-often against header_maxsize. Previously a crafted message could
380 induce a crash of the recive process; now the message is cleanly rejected.
382 JH/42 Bug 2366: Fix the behaviour of the dkim_verify_signers option. It had
383 been totally disabled for all of 4.91. Discovery and fix by "Mad Alex".
389 GF/01 DEFER rather than ERROR on redis cluster MOVED response.
390 When redis_servers is set to a list of > 1 element, and the Redis servers
391 in that list are in cluster configuration, convert the REDIS_REPLY_ERROR
392 case of MOVED into a DEFER case instead, thus moving the query onto the
393 next server in the list. For a cluster of N elements, all N servers must
394 be defined in redis_servers.
396 GF/02 Catch and remove uninitialized value warning in exiqsumm
397 Check for existence of @ARGV before looking at $ARGV[0]
399 JH/01 Replace the store_release() internal interface with store_newblock(),
400 which internalises the check required to safely use the old one, plus
401 the allocate and data copy operations duplicated in both (!) of the
402 extant use locations.
404 JH/02 Disallow '/' characters in queue names specified for the "queue=" ACL
405 modifier. This matches the restriction on the commandline.
407 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
408 Previously only the last row was returned.
410 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
411 we assumed that tags in the header were well-formed, and parsed the
412 element content after inspecting only the first char of the tag.
413 Assumptions at that stage could crash the receive process on malformed
416 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
417 While running the DKIM ACL we operate on the Permanent memory pool so that
418 variables created with "set" persist to the DATA ACL. Also (at any time)
419 DNS lookups that fail create cache records using the Permanent pool. But
420 expansions release any allocations made on the current pool - so a dnsdb
421 lookup expansion done in the DKIM ACL releases the memory used for the
422 DNS negative-cache, and bad things result. Solution is to switch to the
423 Main pool for expansions.
424 While we're in that code, add checks on the DNS cache during store_reset,
425 active in the testsuite.
426 Problem spotted, and debugging aided, by Wolfgang Breyha.
428 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
429 When none of the hosts presented to a transport match an already-open
430 connection, close it and proceed with the list. Previously we would
431 queue the message. Spotted by Lena with Yahoo, probably involving
434 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
435 Previously a spurious "250 OK id=" response was appended to the proper
438 JH/08 The "support for" informational output now, which built with Content
439 Scanning support, has a line for the malware scanner interfaces compiled
440 in. Interface can be individually included or not at build time.
442 JH/09 The "aveserver", "kavdaemon" and "mksd" interfaces are now not included
443 by the template makefile "src/EDITME". The "STREAM" support for an older
444 ClamAV interface method is removed.
446 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
447 rows affected is given instead).
449 JH/11 The runtime Berkeley DB library version is now additionally output by
450 "exim -d -bV". Previously only the compile-time version was shown.
452 JH/12 Bug 2230: Fix cutthrough routing for nonfirst messages in an initiating
453 SMTP connection. Previously, when one had more receipients than the
454 first, an abortive onward connection was made. Move to full support for
455 multiple onward connections in sequence, handling cutthrough connection
456 for all multi-message initiating connections.
458 JH/13 Bug 2229: Fix cutthrough routing for nonstandard port numbers defined by
459 routers. Previously, a multi-recipient message would fail to match the
460 onward-connection opened for the first recipient, and cause its closure.
462 JH/14 Bug 2174: A timeout on connect for a callout was also erroneously seen as
463 a timeout on read on a GnuTLS initiating connection, resulting in the
464 initiating connection being dropped. This mattered most when the callout
465 was marked defer_ok. Fix to keep the two timeout-detection methods
468 JH/15 Relax results from ACL control request to enable cutthrough, in
469 unsupported situations, from error to silently (except under debug)
470 ignoring. This covers use with PRDR, frozen messages, queue-only and
473 HS/01 Fix Buffer overflow in base64d() (CVE-2018-6789)
475 JH/16 Fix bug in DKIM verify: a buffer overflow could corrupt the malloc
476 metadata, resulting in a crash in free().
478 PP/01 Fix broken Heimdal GSSAPI authenticator integration.
479 Broken in f2ed27cf5, missing an equals sign for specified-initialisers.
480 Broken also in d185889f4, with init system revamp.
482 JH/17 Bug 2113: Fix conversation closedown with the Avast malware scanner.
483 Previously we abruptly closed the connection after reading a malware-
484 found indication; now we go on to read the "scan ok" response line,
487 JH/18 Bug 2239: Enforce non-usability of control=utf8_downconvert in the mail
488 ACL. Previously, a crash would result.
490 JH/19 Speed up macro lookups during configuration file read, by skipping non-
491 macro text after a replacement (previously it was only once per line) and
492 by skipping builtin macros when searching for an uppercase lead character.
494 JH/20 DANE support moved from Experimental to mainline. The Makefile control
495 for the build is renamed.
497 JH/21 Fix memory leak during multi-message connections using STARTTLS. A buffer
498 was allocated for every new TLS startup, meaning one per message. Fix
499 by only allocating once (OpenSSL) or freeing on TLS-close (GnuTLS).
501 JH/22 Bug 2236: When a DKIM verification result is overridden by ACL, DMARC
502 reported the original. Fix to report (as far as possible) the ACL
503 result replacing the original.
505 JH/23 Fix memory leak during multi-message connections using STARTTLS under
506 OpenSSL. Certificate information is loaded for every new TLS startup,
507 and the resources needed to be freed.
509 JH/24 Bug 2242: Fix exim_dbmbuild to permit directoryless filenames.
511 JH/25 Fix utf8_downconvert propagation through a redirect router. Previously it
514 JH/26 Bug 2253: For logging delivery lines under PRDR, append the overall
515 DATA response info to the (existing) per-recipient response info for
516 the "C=" log element. It can have useful tracking info from the
517 destination system. Patch from Simon Arlott.
519 JH/27 Bug 2251: Fix ldap lookups that return a single attribute having zero-
520 length value. Previously this would segfault.
522 HS/02 Support Avast multiline protoocol, this allows passing flags to
523 newer versions of the scanner.
525 JH/28 Ensure that variables possibly set during message acceptance are marked
526 dead before release of memory in the daemon loop. This stops complaints
527 about them when the debug_store option is enabled. Discovered specifically
528 for sender_rate_period, but applies to a whole set of variables.
529 Do the same for the queue-runner and queue-list loops, for variables set
530 from spool message files. Do the same for the SMTP per-message loop, for
531 certain variables indirectly set in ACL operations.
533 JH/29 Bug 2250: Fix a longstanding bug in heavily-pipelined SMTP input (such
534 as a multi-recipient message from a mailinglist manager). The coding had
535 an arbitrary cutoff number of characters while checking for more input;
536 enforced by writing a NUL into the buffer. This corrupted long / fast
537 input. The problem was exposed more widely when more pipelineing of SMTP
538 responses was introduced, and one Exim system was feeding another.
539 The symptom is log complaints of SMTP syntax error (NUL chars) on the
540 receiving system, and refused recipients seen by the sending system
541 (propating to people being dropped from mailing lists).
542 Discovered and pinpointed by David Carter.
544 JH/30 The (EXPERIMENTAL_DMARC) variable $dmarc_ar_header is withdrawn, being
545 replaced by the ${authresults } expansion.
547 JH/31 Bug 2257: Fix pipe transport to not use a socket-only syscall.
549 HS/03 Set a handler for SIGTERM and call exit(3) if running as PID 1. This
550 allows proper process termination in container environments.
552 JH/32 Bug 2258: Fix spool_wireformat in combination with LMTP transport.
553 Previously the "final dot" had a newline after it; ensure it is CR,LF.
555 JH/33 SPF: remove support for the "spf" ACL condition outcome values "err_temp"
556 and "err_perm", deprecated since 4.83 when the RFC-defined words
557 "temperror" and "permerror" were introduced.
559 JH/34 Re-introduce enforcement of no cutthrough delivery on transports having
560 transport-filters or DKIM-signing. The restriction was lost in the
561 consolidation of verify-callout and delivery SMTP handling.
562 Extend the restriction to also cover ARC-signing.
564 JH/35 Cutthrough: for a final-dot response timeout (and nonunderstood responses)
565 in defer=pass mode supply a 450 to the initiator. Previously the message
568 PP/02 DANE: add dane_require_tls_ciphers SMTP Transport option; if unset,
569 tls_require_ciphers is used as before.
571 HS/03 Malware Avast: Better match the Avast multiline protocol. Add
572 "pass_unscanned". Only tmpfails from the scanner are written to
573 the paniclog, as they may require admin intervention (permission
574 denied, license issues). Other scanner errors (like decompression
575 bombs) do not cause a paniclog entry.
577 JH/36 Fix reinitialisation of DKIM logging variable between messages.
578 Previously it was possible to log spurious information in receive log
581 JH/37 Bug 2255: Revert the disable of the OpenSSL session caching. This
582 triggered odd behaviour from Outlook Express clients.
584 PP/03 Add util/renew-opendmarc-tlds.sh script for safe renewal of public
587 JH/38 DKIM: accept Ed25519 pubkeys in SubjectPublicKeyInfo-wrapped form,
588 since the IETF WG has not yet settled on that versus the original
589 "bare" representation.
591 JH/39 Fix syslog logging for syslog_timestamp=no and log_selector +millisec.
592 Previously the millisecond value corrupted the output.
593 Fix also for syslog_pid=no and log_selector +pid, for which the pid
594 corrupted the output.
600 JH/01 Rework error string handling in TLS interface so that the caller in
601 more cases is responsible for logging. This permits library-sourced
602 string to be attached to addresses during delivery, and collapses
603 pairs of long lines into single ones.
605 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
606 during configuration. Wildcards are allowed and expanded.
608 JH/02 Rework error string handling in DKIM to pass more info back to callers.
609 This permits better logging.
611 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
612 do not close it down and have the child transport start it up again on
613 the passed-on TCP connection. Instead, proxy the child (and any
614 subsequent ones) for TLS via a unix-domain socket channel. Logging is
615 affected: the continued delivery log lines do not have any DNSSEC, TLS
616 Certificate or OCSP information. TLS cipher information is still logged.
618 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
619 identical IP addresses on different listening ports. Will also affect
622 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
623 add noisy ifdef guards to special-case this sillyness.
624 Patch from Bernd Kuhls.
626 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
627 than 255 are no longer allowed.
629 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
630 Disable the session-cache too, which might reduce our load. Since we
631 currrectly use a new context for every connection, both as server and
632 client, there is no benefit for these.
633 GnuTLS appears to not support tickets server-side by default (we don't
634 call gnutls_session_ticket_enable_server()) but client side is enabled
635 by default on recent versions (3.1.3 +) unless the PFS priority string
638 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
639 <https://reproducible-builds.org/specs/source-date-epoch/>.
641 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
642 the check for any unsuccessful recipients did not notice the limit, and
643 erroneously found still-pending ones.
645 JH/08 Pipeline CHUNKING command and data together, on kernels that support
646 MSG_MORE. Only in-clear (not on TLS connections).
648 JH/09 Avoid using a temporary file during transport using dkim. Unless a
649 transport-filter is involved we can buffer the headers in memory for
650 creating the signature, and read the spool data file once for the
651 signature and again for transmission.
653 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
654 4.77 as the kernel support then wasn't solid, having issues in 64bit
655 mode. Now, it's been long enough. Add support for FreeBSD also.
657 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
658 case where the routing stage had gathered several addresses to send to
659 a host before calling the transport for the first, we previously failed
660 to close down TLS in the old transport process before passing the TCP
661 connection to the new process. The new one sent a STARTTLS command
662 which naturally failed, giving a failed delivery and bloating the retry
663 database. Investigation and fix prototype from Wolfgang Breyha.
665 JH/12 Fix check on SMTP command input synchronisation. Previously there were
666 false-negatives in the check that the sender had not preempted a response
667 or prompt from Exim (running as a server), due to that code's lack of
668 awareness of the SMTP input buffering.
670 PP/04 Add commandline_checks_require_admin option.
671 Exim drops privileges sanely, various checks such as -be aren't a
672 security problem, as long as you trust local users with access to their
673 own account. When invoked by services which pass untrusted data to
674 Exim, this might be an issue. Set this option in main configuration
675 AND make fixes to the calling application, such as using `--` to stop
678 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
679 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
680 responses to those, into a single TLS record each way (this usually means
681 a single packet). As a side issue, smtp_enforce_sync now works on TLS
684 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
685 affects you only if you're dancing at the edge of the param size limits.
686 If you are, and this message makes sense to you, then: raise the
687 configured limit or use OpenSSL 1.1. Nothing we can do for older
690 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
691 cmdline element to get the documented default one. Previously it was
694 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
695 are unsuable for non-message-specific reasons - eg. connection timeout,
696 banner-time rejection.
698 JH/16 Fix logging of delivery remote port, when specified by router, under
701 PP/06 Repair manualroute's ability to take options in any order, even if one
702 is the name of a transport.
705 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
707 JH/17 Change the list-building routines interface to use the expanding-string
708 triplet model, for better allocation and copying behaviour.
710 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
711 Previously it was constructed the first time a possibly-matching string
712 was met in the configuration file input during startup; now it is done
715 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
716 compatible one, to avoid the (poorly documented) possibility of a config
717 file in the working directory redirecting the DB files, possibly correpting
718 some existing file. CVE-2017-10140 assigned for BDB.
720 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
721 cache-hot. Previously, although the result was properly cached, the
722 initial verify call returned a defer.
724 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
725 the main verify for receipient in uncached-mode.
727 JH/22 Retire historical build files to an "unsupported" subdir. These are
728 defined as "ones for which we have no current evidence of testing".
730 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
731 if present. Previously it was ignored.
733 JH/24 Start using specified-initialisers in C structure init coding. This is
734 a C99 feature (it's 2017, so now considered safe).
736 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
737 if was a fixed-sized field and bitmask ops via macros; it is now more
740 PP/07 GitHub PR 56: Apply MariaDB build fix.
741 Patch provided by Jaroslav Škarvada.
743 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
744 during Coverity cleanups [4.87 JH/47]
745 Diagnosis and fix provided by Michael Fischer v. Mollard.
747 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
748 the right size to place the terminating semicolon on its own folded
749 line, the header hash was calculated to an incorrect value thanks to
750 the (relaxed) space the fold became.
752 HS/02 Fix Bug 2130: large writes from the transport subprocess were chunked
753 and confused the parent.
755 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
756 which could crash as a result. This could lead to undeliverable messages.
758 JH/28 Logging: "next input sent too soon" now shows where input was truncated
761 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
762 matters on fast-turnover and PID-randomising systems, which were getting
763 out-of-order delivery.
765 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
766 a possibly-overlapping copy. The symptom was that "Remote host closed
767 connection in response to HELO" was logged instead of the actual 4xx
770 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
771 Previously only that bufferd was discarded, resulting in SYMTP command
774 JH/32 DKIM: when a message has multiple signatures matching an identity given
775 in dkim_verify_signers, run the dkim acl once for each. Previously only
776 one run was done. Bug 2189.
778 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
779 "panic the current process" to "deliberately defer". The panic log is
780 still written with the problem list name; the mail and reject logs now
781 get a temp-reject line for the message that was being handled, saying
782 something like "domains check lookup or other defer". The SMTP 451
783 message is still "Temporary local problem".
785 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
786 A crafted sequence of BDAT commands could result in in-use memory beeing
787 freed. CVE-2017-16943.
789 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
790 from SMTP input. Previously it was always done; now only done for DATA
791 and not BDAT commands. CVE-2017-16944.
793 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
794 to the message (such as an overlong header line). Previously this was
795 not done and we did not exit BDAT mode. Followon from the previous item
796 though a different problem.
802 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
803 than -2003 did; needs libidn2 in addition to libidn.
805 JH/02 The path option on a pipe transport is now expanded before use.
807 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
808 Patch provided by "Björn", documentation fix added too.
810 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
811 missing a wire-to-host endian conversion.
813 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
814 close after a BDAT command line could be taken as a following command,
815 giving a synch failure. Fix by only checking for synch immediately
816 before acknowledging the chunk.
818 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
819 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
820 macro. Patches provided by Josh Soref.
822 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
823 Previously we did not; the RFC seems ambiguous and VRFY is not listed
824 by IANA as a service extension. However, John Klensin suggests that we
827 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
828 the dkim code may be unix-mode line endings rather than smtp wire-format
829 CRLF, so prepend a CR to any bare LF.
831 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
832 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
834 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
835 the first were themselves being wrongly included in the feed into dkim
836 processing; with most chunk sizes in use this resulted in an incorrect
837 body hash calculated value.
839 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
840 DKIM signature block, for verification. Although advised against by
841 standards it is specifically not ruled illegal.
843 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
845 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
846 missing a body hash (the bh= tag).
848 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
849 It seems that HAProxy sends the Proxy Protocol information in clear and
850 only then does a TLS startup, so do the same.
852 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
853 TCP connections (such as for Spamd) unless the daemon successfully set
854 Fast Open mode on its listening sockets. This fixes breakage seen on
855 too-old kernels or those not configured for Fast Open, at the cost of
856 requiring both directions being enabled for TFO, and TFO never being used
857 by non-daemon-related Exim processes.
859 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
860 endings, at least on the first header line. Try to canonify any that get
861 past that check, despite the cost.
863 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
864 now limited to an arbitrary five deep, while parsing addresses with the
865 strip_excess_angle_brackets option enabled.
867 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
868 instead leave the unprompted TLS handshake in socket buffer for the
869 TLS library to consume.
871 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
873 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
875 JH/16 Drop variables when they go out of scope. Memory management drops a whole
876 region in one operation, for speed, and this leaves assigned pointers
877 dangling. Add checks run only under the testsuite which checks all
878 variables at a store-reset and panics on a dangling pointer; add code
879 explicitly nulling out all the variables discovered. Fixes one known
880 bug: a transport crash, where a dangling pointer for $sending_ip_address
881 originally assigned in a verify callout, is re-used.
883 PP/06 Drop '.' from @INC in various Perl scripts.
885 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
887 PP/08 Reduce a number of compilation warnings under clang; building with
888 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
889 should be warning-free.
891 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
893 HS/01 Fix portability problems introduced by PP/08 for platforms where
894 realloc(NULL) is not equivalent to malloc() [SunOS et al].
896 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
897 chunk. This allows us to accept broken chunked messages. We need a more
898 general solution here.
900 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
901 already-broken messages in the queue.
903 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
905 JH/19 Fix reference counting bug in routing-generated-address tracking.
911 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
912 supports it and a size is available (ie. the sending peer gave us one).
914 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
915 years of being deprecated). The replacements are the ACLs
916 acl_smtp_mime and acl_not_smtp_mime.
918 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
919 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
920 or even an in-clear connection were permitted. Now, if the host lookup
921 was dnssec and dane was requested then the host is only used if the
922 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
923 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
924 if one fails this test.
925 This means that a poorly-configured remote DNS will make it incommunicado;
926 but it protects against a DNS-interception attack on it.
928 JH/04 Bug 1810: make continued-use of an open smtp transport connection
929 non-noisy when a race steals the message being considered.
931 JH/05 If main configuration option tls_certificate is unset, generate a
932 self-signed certificate for inbound TLS connections.
934 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
935 in rewrites and routers.
937 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
938 and logged a warning sing 4.83; now they are a configuration file error.
940 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
941 (lacking @domain). Apply the same qualification processing as RCPT.
943 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
945 JH/10 Support ${sha256:} applied to a string (as well as the previous
948 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
949 a cutthrough deliver is pending, as we always want to make a connection.
950 This also avoids re-routing the message when later placing the cutthrough
951 connection after a verify cache hit.
952 Do not update it with the verify result either.
954 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
955 when routing results in more than one destination address.
957 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
958 signing (which inhibits the cutthrough capability). Previously only
959 the presence of an option was tested; now an expansion evaluating as
960 empty is permissible (obviously it should depend only on data available
961 when the cutthrough connection is made).
963 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
964 the relevant preceding SMTP command did not note the pipelining mode.
966 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
967 Previously they were not counted.
969 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
970 as one having no matching records. Previously we deferred the message
971 that needed the lookup.
973 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
974 distinguished as "(=".
976 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
977 for missing MX records. Previously it only worked for missing A records.
979 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
981 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
982 after the data-go-ahead and data-ack. Patch from Jason Betts.
984 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
985 even for a "none" policy. Patch from Tony Meyer.
987 JH/22 Fix continued use of a connection for further deliveries. If a port was
988 specified by a router, it must also match for the delivery to be
991 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
992 When one of the recipients of a message was unsuitable for the connection
993 (has no matching addresses), we lost track of needing to mark it
994 deferred. As a result mail would be lost.
996 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
998 JH/25 Decoding ACL controls is now done using a binary search; the source code
999 takes up less space and should be simpler to maintain. Merge the ACL
1000 condition decode tables also, with similar effect.
1002 JH/26 Fix problem with one_time used on a redirect router which returned the
1003 parent address unchanged. A retry would see the parent address marked as
1004 delivered, so not attempt the (identical) child. As a result mail would
1007 JH/27 Fix a possible security hole, wherein a process operating with the Exim
1008 UID can gain a root shell. Credit to http://www.halfdog.net/ for
1009 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
1012 JH/28 Enable {spool,log} filesystem space and inode checks as default.
1013 Main config options check_{log,spool}_{inodes,space} are now
1014 100 inodes, 10MB unless set otherwise in the configuration.
1016 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
1017 Previously it only applied to the main-section connection policy
1020 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
1022 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
1023 by me. Added RFC7919 DH primes as an alternative.
1025 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
1026 are not in the system include path.
1028 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
1029 GnuTLS, when a session startup failed (eg because the client disconnected)
1030 Exim did stdio operations after fclose. This was exposed by a recent
1031 change which nulled out the file handle after the fclose.
1033 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
1034 signed directly by the cert-signing cert, rather than an intermediate
1035 OCSP-signing cert. This is the model used by LetsEncrypt.
1037 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
1039 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
1040 an incoming connection.
1042 HS/02 Bug 1802: Do not half-close the connection after sending a request
1045 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
1046 fallback to "prime256v1".
1048 JH/34 SECURITY: Use proper copy of DATA command in error message.
1049 Could leak key material. Remotely exploitable. CVE-2016-9963.
1055 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
1056 and 3.4.4 - once the server is enabled to respond to an OCSP request
1057 it does even when not requested, resulting in a stapling non-aware
1058 client dropping the TLS connection.
1060 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
1061 support variable-length bit vectors. No functional change.
1063 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
1064 The I= interface field on outgoing lines is now after the H= remote
1065 host field, same as incoming lines. There is a separate
1066 outgoing_interface log selector which allows you to disable the
1069 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
1070 If not running log_selector +smtp_connection the mainlog would be held
1071 open indefinitely after a "too many connections" event, including to a
1072 deleted file after a log rotate. Leave the per net connection logging
1073 leaving it open for efficiency as that will be quickly detected by the
1074 check on the next write.
1076 HS/01 Bug 1671: Fix post transport crash.
1077 Processing the wait-<transport> messages could crash the delivery
1078 process if the message IDs didn't exist for some reason. When
1079 using 'split_spool_directory=yes' the construction of the spool
1080 file name failed already, exposing the same netto behaviour.
1082 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
1083 mime_regex ACL conditions.
1085 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
1086 to DSN fail messages (bounces): remote IP, remote greeting, remote response
1087 to HELO, local diagnostic string.
1089 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
1090 log line to debug. Even when configured with a tls authenticator many
1091 client connections are expected to not authenticate in this way, so
1092 an authenticate fail is not an error.
1094 HS/02 Add the Exim version string to the process info. This way exiwhat
1095 gives some more detail about the running daemon.
1097 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
1098 matter for fast-change records such as DNSBLs.
1100 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
1101 retry record, even if constant. There may be multiple transports with
1102 different interface settings and the retry behaviour needs to be kept
1105 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
1107 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
1109 JH/10 Bug 840: fix log_defer_output option of pipe transport
1111 JH/11 Bug 830: use same host for all RCPTS of a message, even under
1112 hosts_randomize. This matters a lot when combined with mua_wrapper.
1114 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
1115 ${quote_pgsql:<string>} operator.
1117 JH/13 Bug 1708: avoid misaligned access in cached lookup.
1119 JH/14 Change header file name for freeradius-client. Relevant if compiling
1120 with Radius support; from the Gentoo tree and checked under Fedora.
1122 JH/15 Bug 1712: Introduce $prdr_requested flag variable
1124 JH/16 Bug 1714: Permit an empty string as expansion result for transport
1125 option transport_filter, meaning no filtering.
1127 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
1129 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
1130 defaults to "*" (all hosts). The variable is now available when not built
1131 with TLS, default unset, mainly to enable keeping the testsuite sane.
1132 If a server certificate is not supplied (via tls_certificate) an error is
1133 logged, and clients will find TLS connections fail on startup. Presumably
1134 they will retry in-clear.
1135 Packagers of Exim are strongly encouraged to create a server certificate
1136 at installation time.
1138 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
1139 with the $config_file variable.
1141 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
1142 in transport context, after the attempt, and per-recipient. The latter type
1143 is per host attempted. The event data is the error message, and the errno
1144 information encodes the lookup type (A vs. MX) used for the (first) host,
1145 and the trailing two digits of the smtp 4xx response.
1147 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
1148 to write to mainlog (or rejectlog, paniclog) in the window between file
1149 creation and permissions/ownership being changed. Particularly affects
1150 installations where exicyclog is run as root, rather than exim user;
1151 result is that the running daemon panics and dies.
1153 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
1155 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
1156 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
1157 "pri" and "weight". Note that the previous implicit priority given by the
1158 list order is no longer honoured.
1160 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
1161 for DKIM processing.
1163 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
1164 by defining SUPPORT_SOCKS.
1166 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
1167 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
1168 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
1169 variables are renamed to proxy_{local,external}_{address,port}.
1171 JH/27 Move Internationalisation support from Experimental to mainline, enabled
1172 for a build by defining SUPPORT_I18N
1174 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
1175 of the query string, and make ${quote_redis:} do that quoting.
1177 JH/29 Move Events support from Experimental to mainline, enabled by default
1178 and removable for a build by defining DISABLE_EVENT.
1180 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
1182 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
1183 cached by the daemon.
1185 JH/32 Move Redis support from Experimental to mainline, enabled for a build
1186 by defining LOOKUP_REDIS. The libhiredis library is required.
1188 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
1189 keys are given for lookup.
1191 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
1192 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
1193 only supported when built with TLS support. The PolarSSL SHA routines
1194 are still used when the TLS library is too old for convenient support.
1196 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
1197 openssl_options), for security. OpenSSL forces this from version 1.1.0
1198 server-side so match that on older versions.
1200 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
1201 allocation for $value could be released as the expansion processing
1202 concluded, but leaving the global pointer active for it.
1204 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
1205 and to use the domains and local_parts ACL conditions.
1207 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
1208 incorrectly not doubled on cutthrough transmission, hence seen as a
1209 body-termination at the receiving system - resulting in truncated mails.
1210 Commonly the sender saw a TCP-level error, and retransmitted the message
1211 via the normal store-and-forward channel. This could result in duplicates
1212 received - but deduplicating mailstores were liable to retain only the
1213 initial truncated version.
1215 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
1217 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
1219 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
1220 we're in there, support oversigning also; bug 1309.
1222 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
1224 HS/04 Add support for keep_environment and add_environment options.
1226 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
1227 either intentional arithmetic overflow during PRNG, or testing config-
1230 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
1231 delivery resulted in actual delivery. Cancel cutthrough before DATA
1234 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
1235 rejects a recipient: pass the reject to the originator.
1237 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
1238 Many were false-positives and ignorable, but it's worth fixing the
1241 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
1242 for the new environment-manipulation done at startup. Move the routines
1243 from being local to tls.c to being global via the os.c file.
1245 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
1246 an extract embedded as result-arg for a map, the first arg for extract
1247 is unavailable so we cannot tell if this is a numbered or keyed
1248 extraction. Accept either.
1254 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
1257 JH/02 The smtp transport option "multi_domain" is now expanded.
1259 JH/03 The smtp transport now requests PRDR by default, if the server offers
1262 JH/04 Certificate name checking on server certificates, when exim is a client,
1263 is now done by default. The transport option tls_verify_cert_hostnames
1264 can be used to disable this per-host. The build option
1265 EXPERIMENTAL_CERTNAMES is withdrawn.
1267 JH/05 The value of the tls_verify_certificates smtp transport and main options
1268 default to the word "system" to access the system default CA bundle.
1269 For GnuTLS, only version 3.0.20 or later.
1271 JH/06 Verification of the server certificate for a TLS connection is now tried
1272 (but not required) by default. The verification status is now logged by
1273 default, for both outbound TLS and client-certificate supplying inbound
1276 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
1279 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
1280 Status Notification (bounce) messages are now MIME format per RFC 3464.
1281 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
1282 under the control of the dsn_advertise_hosts option, and routers may
1283 have a dsn_lasthop option.
1285 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
1286 default, modifiable by a malware= option. The list separator for
1287 the options can now be changed in the usual way. Bug 68.
1289 JH/10 The smtp_receive_timeout main option is now expanded before use.
1291 JH/11 The incoming_interface log option now also enables logging of the
1292 local interface on delivery outgoing connections.
1294 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
1295 if the interface and destination host and port all match.
1297 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
1300 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
1301 Patch from Andrew Lewis.
1303 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
1304 now supports optional time-restrictions, weighting, and priority
1305 modifiers per server. Patch originally by <rommer@active.by>.
1307 JH/16 The spamd_address main option now supports a mixed list of local
1308 and remote servers. Remote servers can be IPv6 addresses, and
1309 specify a port-range.
1311 JH/17 Bug 68: The spamd_address main option now supports an optional
1312 timeout value per server.
1314 JH/18 Bug 1581: Router and transport options headers_add/remove can
1315 now have the list separator specified.
1317 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
1320 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
1323 JH/21 Support for the A6 type of dns record is withdrawn.
1325 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
1326 rather than the verbs used.
1328 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
1329 from 255 to 1024 chars.
1331 JH/24 Verification callouts now attempt to use TLS by default.
1333 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
1334 are generic router options now. The defaults didn't change.
1336 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
1337 Original patch from Alexander Shikoff, worked over by JH.
1339 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
1340 files if ZCAT_COMMAND is not executable.
1342 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
1344 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
1346 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
1347 Normally benign, it bites when the pair was led to by a CNAME;
1348 modern usage is to not canonicalize the domain to a CNAME target
1349 (and we were inconsistent anyway for A-only vs AAAA+A).
1351 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
1353 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
1354 when evaluating $sender_host_dnssec.
1356 JH/31 Check the HELO verification lookup for DNSSEC, adding new
1357 $sender_helo_dnssec variable.
1359 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
1361 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
1363 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
1365 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
1366 documented as working, but never had. Support all but $spam_report.
1368 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
1369 added for tls authenticator.
1371 HS/03 Add perl_taintmode main config option
1377 TL/01 When running the test suite, the README says that variables such as
1378 no_msglog_check are global and can be placed anywhere in a specific
1379 test's script, however it was observed that placement needed to be near
1380 the beginning for it to behave that way. Changed the runtest perl
1381 script to read through the entire script once to detect and set these
1382 variables, reset to the beginning of the script, and then run through
1383 the script parsing/test process like normal.
1385 TL/02 The BSD's have an arc4random API. One of the functions to induce
1386 adding randomness was arc4random_stir(), but it has been removed in
1387 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
1388 function when detected.
1390 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
1391 cause callback expansion.
1393 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
1394 syntax errors in an expansion can be treated as a string instead of
1395 logging or causing an error, due to the internal use of bool_lax
1396 instead of bool when processing it.
1398 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
1399 server certificates when making smtp deliveries.
1401 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
1403 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
1405 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
1407 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
1408 Merged patch from Sebastian Wiedenroth.
1410 JH/05 Fix results-pipe from transport process. Several recipients, combined
1411 with certificate use, exposed issues where response data items split
1412 over buffer boundaries were not parsed properly. This eventually
1413 resulted in duplicates being sent. This issue only became common enough
1414 to notice due to the introduction of connection certificate information,
1415 the item size being so much larger. Found and fixed by Wolfgang Breyha.
1417 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
1418 size buffer was used, resulting in syntax errors when an expansion
1421 JH/07 Add support for directories of certificates when compiled with a GnuTLS
1422 version 3.3.6 or later.
1424 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
1425 is EXPERIMENTAL_EVENT, the main-configuration and transport options
1426 both become "event_action", the variables become $event_name, $event_data
1427 and $event_defer_errno. There is a new variable $verify_mode, usable in
1428 routers, transports and related events. The tls:cert event is now also
1429 raised for inbound connections, if the main configuration event_action
1432 TL/06 In test suite, disable OCSP for old versions of openssl which contained
1433 early OCSP support, but no stapling (appears to be less than 1.0.0).
1435 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
1436 server certificate names available under the smtp transport option
1437 "tls_verify_cert_hostname" now do not permit multi-component wildcard
1440 JH/10 Time-related extraction expansions from certificates now use the main
1441 option "timezone" setting for output formatting, and are consistent
1442 between OpenSSL and GnuTLS compilations. Bug 1541.
1444 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
1445 encoded parameter in the incoming message. Bug 1558.
1447 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
1448 include certificate info, eximon was claiming there were spoolfile
1451 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
1453 JH/14 Log delivery-related information more consistently, using the sequence
1454 "H=<name> [<ip>]" wherever possible.
1456 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
1457 are problematic for Debian distribution, omit them from the release
1460 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
1462 JH/16 Fix string representation of time values on 64bit time_t architectures.
1465 JH/17 Fix a null-indirection in certextract expansions when a nondefault
1466 output list separator was used.
1471 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
1472 checkers that were complaining about end of non-void function with no
1475 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
1476 This was a regression introduced in 4.83 by another bugfix.
1478 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1480 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1481 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1487 TF/01 Correctly close the server side of TLS when forking for delivery.
1489 When a message was received over SMTP with TLS, Exim failed to clear up
1490 the incoming connection properly after forking off the child process to
1491 deliver the message. In some situations the subsequent outgoing
1492 delivery connection happened to have the same fd number as the incoming
1493 connection previously had. Exim would try to use TLS and fail, logging
1494 a "Bad file descriptor" error.
1496 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1497 utilities have not been installed.
1499 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1500 temporary space as the ACL may create new global variables.
1502 TL/01 LDAP support uses per connection or global context settings, depending
1503 upon the detected version of the libraries at build time.
1505 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1506 to extract and use the src ip:port in logging and expansions as if it
1507 were a direct connection from the outside internet. PPv2 support was
1508 updated based on HAProxy spec change in May 2014.
1510 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1512 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1513 Properly escape header and check for NULL return.
1515 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1518 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1520 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1521 characters in header names, implemented as a verify condition.
1522 Contributed by Michael Fischer v. Mollard.
1524 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1525 results permerror and temperror. Previous values are deprecated but
1526 still accepted. In a future release, err_perm and err_temp will be
1527 completely removed, which will be a backward incompatibility if the
1528 ACL tests for either of these two old results. Patch contributed by
1529 user bes-internal on the mailing list.
1531 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1533 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1534 selectors, in both main and reject logs.
1536 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1539 JH/07 Add malware type "sock" for talking to simple daemon.
1541 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1543 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1544 routers/transports under cutthrough routing.
1546 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1547 numbers. Touch up "bool" conditional to keep the same definition.
1549 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1551 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1553 JH/12 Expand items in router/transport headers_add or headers_remove lists
1554 individually rather than the list as a whole. Bug 1452.
1556 Required for reasonable handling of multiple headers_ options when
1557 they may be empty; requires that headers_remove items with embedded
1558 colons must have them doubled (or the list-separator changed).
1560 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1561 view the policy declared in the DMARC record. Currently, $dmarc_status
1562 is a combined value of both the record presence and the result of the
1565 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1567 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1568 dnslookup router and the smtp transport (applying to the forward
1571 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1572 of ldap servers used for a specific lookup. Patch provided by Heiko
1575 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1576 New variable $lookup_dnssec_authenticated for observability.
1578 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1579 Patch submitted by Lars Timman.
1581 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1583 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1584 Requires trusted mode and valid format message id, aborts otherwise.
1585 Patch contributed by Heiko Schlichting.
1587 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1588 certextract with support for various fields. Bug 1358.
1590 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1591 is requested by default, modifiable by smtp transport option
1594 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1595 operate on certificate variables to give certificate fingerprints
1596 Also new ${sha256:cert_variable}.
1598 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1600 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1601 Christian Aistleitner.
1603 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1605 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1606 file. Patch from Wolfgang Breyha.
1608 JH/25 Expand the coverage of the delivery $host and $host_address to
1609 client authenticators run in verify callout. Bug 1476.
1611 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1612 align with daemon_smtp_ports. Bug 72.
1614 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1615 support and error reporting did not work properly.
1617 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1618 and is readable. Patch from Andrew Colin Kissa.
1620 TL/14 Enhance documentation of ${run expansion and how it parses the
1621 commandline after expansion, particularly in the case when an
1622 unquoted variable expansion results in an empty value.
1624 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1626 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1627 by renaming away from it. Addresses GH issue 15, reported by
1630 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1632 TL/15 SECURITY: prevent double expansion in math comparison functions
1633 (can expand unsanitized data). Not remotely exploitable.
1640 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1642 PP/02 Make -n do something, by making it not do something.
1643 When combined with -bP, the name of an option is not output.
1645 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1648 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1649 $sender_host_name and config options to manage this, and basic check
1652 PP/05 DSCP support for outbound connections and control modifier for inbound.
1654 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1655 (Only plugin which currently uses this is kerberos4, which nobody should
1656 be using, but we should make it available and other future plugins might
1657 conceivably use it, even though it would break NAT; stuff *should* be
1658 using channel bindings instead).
1660 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1661 name; added for Sendmail compatibility; requires admin caller.
1662 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1663 just ignore it); requires trusted caller.
1664 Also parse but ignore: -Ac -Am -X<logfile>
1667 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1669 TL/02 Add +smtp_confirmation as a default logging option.
1671 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1672 Patch by Magnus Holmgren from 2007-02-20.
1674 TL/04 Bugzilla 1281 - Spec typo.
1675 Bugzilla 1283 - Spec typo.
1676 Bugzilla 1290 - Spec grammar fixes.
1678 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1680 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1682 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1683 mailing list by by Dmitry Isaikin.
1685 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1687 JH/02 Support "G" suffix to numbers in ${if comparisons.
1689 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1691 NM/01 Bugzilla 1197 - Spec typo
1692 Bugzilla 1196 - Spec examples corrections
1694 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1696 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1697 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1700 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1701 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1703 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1704 "acl {{name}{arg}...}", and optional args on acl condition
1707 JH/05 Permit multiple router/transport headers_add/remove lines.
1709 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1711 JH/07 Avoid using a waiting database for a single-message-only transport.
1712 Performance patch from Paul Fisher. Bugzilla 1262.
1714 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1717 JH/09 Add $headers_added variable, with content from use of ACL modifier
1718 add_header (but not yet added to the message). Bugzilla 199.
1720 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1721 Pulled from Bugzilla 817 by Wolfgang Breyha.
1723 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1725 (nb: this is the same fix as in Exim 4.80.1)
1727 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1730 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1733 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1735 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1736 advertise SMTP AUTH mechanism to us, instead of a generic
1737 protocol violation error. Also, make Exim more robust to bad
1738 data from the Dovecot auth socket.
1740 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1742 When a queue runner is handling a message, Exim first routes the
1743 recipient addresses, during which it prunes them based on the retry
1744 hints database. After that it attempts to deliver the message to
1745 any remaining recipients. It then updates the hints database using
1748 So if a recipient address works intermittently, it can get repeatedly
1749 deferred at routing time. The retry hints record remains fresh so the
1750 address never reaches the final cutoff time.
1752 This is a fairly common occurrence when a user is bumping up against
1753 their storage quota. Exim had some logic in its local delivery code
1754 to deal with this. However it did not apply to per-recipient defers
1755 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1757 This change adds a proper retry rule check during routing so that the
1758 final cutoff time is checked against the message's age. We only do
1759 this check if there is an address retry record and there is not a
1760 domain retry record; this implies that previous attempts to handle
1761 the address had the retry_use_local_parts option turned on. We use
1762 this as an approximation for the destination being like a local
1763 delivery, as in LMTP.
1765 I suspect this new check makes the old local delivery cutoff check
1766 redundant, but I have not verified this so I left the code in place.
1768 TF/02 Correct gecos expansion when From: is a prefix of the username.
1770 Test 0254 submits a message to Exim with the header
1774 When I ran the test suite under the user fanf2, Exim expanded
1775 the header to contain my full name, whereas it should have added
1776 a Resent-Sender: header. It erroneously treats any prefix of the
1777 username as equal to the username.
1779 This change corrects that bug.
1781 GF/01 DCC debug and logging tidyup
1782 Error conditions log to paniclog rather than rejectlog.
1783 Debug lines prefixed by "DCC: " to remove any ambiguity.
1785 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1787 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1788 Bug spotted by Jeremy Harris; was flawed since initial commit.
1789 Would have resulted in OCSP responses post-SNI triggering an Exim
1790 NULL dereference and crash.
1792 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1794 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1795 Bug detection, analysis and fix by Samuel Thibault.
1796 Bugzilla 1331, Debian bug #698092.
1798 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1800 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1801 Server implementation by Todd Lyons, client by JH.
1802 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1803 config variable "prdr_enable" controls whether the server
1804 advertises the facility. If the client requests PRDR a new
1805 acl_data_smtp_prdr ACL is called once for each recipient, after
1806 the body content is received and before the acl_smtp_data ACL.
1807 The client is controlled by both of: a hosts_try_prdr option
1808 on the smtp transport, and the server advertisement.
1809 Default client logging of deliveries and rejections involving
1810 PRDR are flagged with the string "PRDR".
1812 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1813 fclose(). Diagnosis by Todd Lyons.
1815 PP/17 Update configure.default to handle IPv6 localhost better.
1816 Patch by Alain Williams (plus minor tweaks).
1819 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1820 This is now consistent with GnuTLS, and is now documented: the
1821 previous undocumented portable approach to treating the option as
1822 unset was to force an expansion failure. That still works, and
1823 an empty string is now equivalent.
1825 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1826 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1827 not performing validation itself.
1829 PP/20 Added force_command boolean option to pipe transport.
1830 Patch from Nick Koston, of cPanel Inc.
1832 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1835 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1837 PP/21 Fix eximon continuous updating with timestamped log-files.
1838 Broken in a format-string cleanup in 4.80, missed when I repaired the
1839 other false fix of the same issue.
1840 Report and fix from Heiko Schlichting.
1843 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1844 Report from Prashanth Katuri.
1846 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1847 It's SecureTransport, so affects any MacOS clients which use the
1848 system-integrated TLS libraries, including email clients.
1850 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1851 using a MIME ACL for non-SMTP local injection.
1852 Report and assistance in diagnosis by Warren Baker.
1854 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1856 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1858 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1859 last id that failed so it may be referenced in subsequent ACL's.
1861 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1864 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1865 ldap library initialization, allowing self-signed CA's to be
1866 used. Also properly sets require_cert option later in code by
1867 using NULL (global ldap config) instead of ldap handle (per
1868 session). Bug diagnosis and testing by alxgomz.
1870 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1871 the src/util/ subdirectory.
1873 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1874 renamed to Transport Post Delivery Action by Jeremy Harris, as
1877 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1878 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1879 redis_servers = needs to be configured which will be used by the redis
1880 lookup. Patch from Warren Baker, of The Packet Hub.
1882 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1884 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1885 hostname or reverse DNS when processing a host list. Used suggestions
1886 from multiple comments on this bug.
1888 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1890 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1891 Missed a few lines, added it to make the runtest require no keyboard
1894 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1895 contains upper case chars. Make router use caseful_local_part.
1897 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1898 support when GnuTLS has been built with p11-kit.
1904 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1906 This, or similar/improved, will also be change PP/11 of 4.82.
1912 PP/01 Handle short writes when writing local log-files.
1913 In practice, only affects FreeBSD (8 onwards).
1914 Bugzilla 1053, with thanks to Dmitry Isaikin.
1916 NM/01 Bugzilla 949 - Documentation tweak
1918 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1921 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1923 PP/02 Implemented gsasl authenticator.
1925 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1927 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1928 `pkg-config foo` for cflags/libs.
1930 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1931 with rest of GSASL and with heimdal_gssapi.
1933 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1934 `pkg-config foo` for cflags/libs for the TLS implementation.
1936 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1937 properties get this fed in as external SSF. A number of robustness
1938 and debugging improvements to the cyrus_sasl authenticator.
1940 PP/08 cyrus_sasl server now expands the server_realm option.
1942 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1943 Patch by Jeremy Harris.
1945 PP/10 Added dbmjz lookup type.
1947 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1949 PP/12 MAIL args handles TAB as well as SP, for better interop with
1950 non-compliant senders.
1951 Analysis and variant patch by Todd Lyons.
1953 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1954 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1955 Patch from Dirk Mueller <dmueller@suse.com>
1957 PP/13 tls_peerdn now print-escaped for spool files.
1958 Observed some $tls_peerdn in wild which contained \n, which resulted
1959 in spool file corruption.
1961 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1962 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1963 or write after TLS renegotiation, which otherwise led to messages
1966 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1967 as a tracking header (ie: a signed header comes before the signature).
1968 Patch from Wolfgang Breyha.
1970 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1971 comma-sep list; embedded commas doubled.
1973 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1975 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1977 Report and patch from Dmitry Banschikov.
1979 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1980 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1981 failures. We appear to now support TLS1.1+ with Exim.
1983 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1984 lets Exim select keys and certificates based upon TLS SNI from client.
1985 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1986 before an outbound SMTP session. New log_selector, +tls_sni.
1988 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1989 NULL dereference. Report and patch from Alun Jones.
1991 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1992 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1993 Not seeing resolver debug output on NetBSD, but suspect this is a
1994 resolver implementation change.
1996 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1997 Left warnings. Added "eximon gdb" invocation mode.
1999 PP/21 Defaulting "accept_8bitmime" to true, not false.
2001 PP/22 Added -bw for inetd wait mode support.
2003 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
2004 locate the relevant includes and libraries. Made this the default.
2006 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
2007 Bugzilla 1246, report and most of solution from Tomasz Kusy.
2009 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
2010 This may cause build issues on older platforms.
2012 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
2013 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
2014 gnutls_require_mac & gnutls_require_protocols (no longer supported).
2015 Added SNI support via GnuTLS too.
2016 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
2018 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
2020 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
2021 Applied second patch from Janne, implementing suggestion to default
2022 multiple-strings-in-record handling to match SPF spec.
2024 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
2026 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
2027 read-only, out of scope).
2028 Patch from Wolfgang Breyha, report from Stuart Northfield.
2030 PP/29 Fix three issues highlighted by clang analyser static analysis.
2031 Only crash-plausible issue would require the Cambridge-specific
2032 iplookup router and a misconfiguration.
2033 Report from Marcin Mirosław.
2035 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
2037 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
2038 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
2039 As part of this, removing so much warning spew let me fix some minor
2040 real issues in debug logging.
2042 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
2043 assignment on my part. Fixed.
2045 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
2046 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
2047 Janne Snabb (who went above and beyond: thank you).
2049 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
2050 string otherwise requires a connection and a bunch more work and it's
2051 relatively easy to get wrong. Should also expose TLS library linkage
2054 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
2055 64-bit ${eval} (JH/03).
2057 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
2058 GNU libc to support some of the 64-bit stuff, should not lead to
2059 conflicts. Defined before os.h is pulled in, so if a given platform
2060 needs to override this, it can.
2062 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
2063 protection layer was required, which is not implemented.
2064 Bugzilla 1254, patch from Wolfgang Breyha.
2066 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
2067 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
2068 tls_dhparam take prime identifiers. Also unbreak combination of
2069 OpenSSL+DH_params+TLSSNI.
2071 PP/39 Disable SSLv2 by default in OpenSSL support.
2077 PP/01 Solaris build fix for Oracle's LDAP libraries.
2078 Bugzilla 1109, patch from Stephen Usher.
2080 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
2082 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
2085 TF/02 Fix a couple more cases where we did not log the error message
2086 when unlink() failed. See also change 4.74-TF/03.
2088 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
2089 lock up or crash if it happened to be inside a call to libc when it
2090 got a SIGUSR1 from exiwhat.
2092 The SIGUSR1 handler appends the current process status to the process
2093 log which is later printed by exiwhat. It used to use the general
2094 purpose logging code to do this, but several functions it calls are
2095 not safe for signals.
2097 The new output code in the SIGUSR1 handler is specific to the process
2098 log, and simple enough that it's easy to inspect for signal safety.
2099 Removing some special cases also simplifies the general logging code.
2100 Removing the spurious timestamps from the process log simplifies
2103 TF/04 Improved ratelimit ACL condition.
2105 The /noupdate option has been deprecated in favour of /readonly which
2106 has clearer semantics. The /leaky, /strict, and /readonly update modes
2107 are mutually exclusive. The update mode is no longer included in the
2108 database key; it just determines when the database is updated. (This
2109 means that when you upgrade Exim will forget old rate measurements.)
2111 Exim now checks that the per_* options are used with an update mode that
2112 makes sense for the current ACL. For example, when Exim is processing a
2113 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
2114 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
2115 must specify per_mail/readonly. If you omit the update mode it defaults to
2116 /leaky where that makes sense (as before) or /readonly where required.
2118 The /noupdate option is now undocumented but still supported for
2119 backwards compatibility. It is equivalent to /readonly except that in
2120 ACLs where /readonly is required you may specify /leaky/noupdate or
2121 /strict/noupdate which are treated the same as /readonly.
2123 A useful new feature is the /count= option. This is a generalization
2124 of the per_byte option, so that you can measure the throughput of other
2125 aggregate values. For example, the per_byte option is now equivalent
2126 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
2128 The per_rcpt option has been generalized using the /count= mechanism
2129 (though it's more complicated than the per_byte equivalence). When it is
2130 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
2131 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
2132 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
2133 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
2134 Note that using per_rcpt with a non-readonly update mode in more than
2135 one ACL will cause the recipients to be double-counted. (The per_mail
2136 and per_byte options don't have this problem.)
2138 The handling of very low rates has changed slightly. If the computed rate
2139 is less than the event's count (usually one) then this event is the first
2140 after a long gap. In this case the rate is set to the same as this event's
2141 count, so that the first message of a spam run is counted properly.
2143 The major new feature is a mechanism for counting the rate of unique
2144 events. The new per_addr option counts the number of different
2145 recipients that someone has sent messages to in the last time period. It
2146 behaves like per_rcpt if all the recipient addresses are different, but
2147 duplicate recipient addresses do not increase the measured rate. Like
2148 the /count= option this is a general mechanism, so the per_addr option
2149 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
2150 example, measure the rate that a client uses different sender addresses
2151 with the options per_mail/unique=$sender_address. There are further
2152 details in the main documentation.
2154 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
2156 TF/06 Removed a few PCRE remnants.
2158 TF/07 Automatically extract Exim's version number from tags in the git
2159 repository when doing development or release builds.
2161 PP/02 Raise smtp_cmd_buffer_size to 16kB.
2162 Bugzilla 879. Patch from Paul Fisher.
2164 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
2165 Heavily based on revision 40f9a89a from Simon Arlott's tree.
2168 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
2170 PP/05 Variable $av_failed, true if the AV scanner deferred.
2171 Bugzilla 1078. Patch from John Horne.
2173 PP/06 Stop make process more reliably on build failure.
2174 Bugzilla 1087. Patch from Heiko Schlittermann.
2176 PP/07 Make maildir_use_size_file an _expandable_ boolean.
2177 Bugzilla 1089. Patch from Heiko Schlittermann.
2179 PP/08 Handle ${run} returning more data than OS pipe buffer size.
2180 Bugzilla 1131. Patch from Holger Weiß.
2182 PP/09 Handle IPv6 addresses with SPF.
2183 Bugzilla 860. Patch from Wolfgang Breyha.
2185 PP/10 GnuTLS: support TLS 1.2 & 1.1.
2187 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
2190 PP/11 match_* no longer expand right-hand-side by default.
2191 New compile-time build option, EXPAND_LISTMATCH_RHS.
2192 New expansion conditions, "inlist", "inlisti".
2194 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
2196 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
2198 PP/14 fix log_write() format string regression from TF/03.
2199 Bugzilla 1152. Patch from Dmitry Isaikin.
2205 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
2207 PP/02 Harmonised TLS library version reporting; only show if debugging.
2208 Layout now matches that introduced for other libraries in 4.74 PP/03.
2210 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
2212 PP/04 New "dns_use_edns0" global option.
2214 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
2217 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
2218 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
2220 TK/01 Updated PolarSSL code to 0.14.2.
2221 Bugzilla 1097. Patch from Andreas Metzler.
2223 PP/07 Catch divide-by-zero in ${eval:...}.
2224 Fixes bugzilla 1102.
2226 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
2229 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
2230 format-string attack -- SECURITY: remote arbitrary code execution.
2232 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
2233 time unintentionally subject to list matching rules, letting the header
2234 cause arbitrary Exim lookups (of items which can occur in lists, *not*
2235 arbitrary string expansion). This allowed for information disclosure.
2237 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
2238 INT_MIN/-1 -- value coerced to INT_MAX.
2244 NM/01 Workaround for PCRE version dependency in version reporting
2247 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
2248 This fixes portability to compilers other than gcc, notably
2249 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
2251 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
2252 makefiles for portability to HP-UX and POSIX correctness.
2254 PP/01 Permit LOOKUP_foo enabling on the make command-line.
2255 Also via indented variable definition in the Makefile.
2256 (Debugging by Oliver Heesakkers).
2258 PP/02 Restore caching of spamd results with expanded spamd_address.
2259 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
2261 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
2262 Improves build reliability. Fix from: Frank Elsner
2264 NM/02 Fix wide character breakage in the rfc2047 coding
2265 Fixes bug 1064. Patch from Andrey N. Oktyabrski
2267 NM/03 Allow underscore in dnslist lookups
2268 Fixes bug 1026. Patch from Graeme Fowler
2270 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
2271 Code patches from Adam Ciarcinski of NetBSD.
2273 NM/04 Fixed exiqgrep to cope with mailq missing size issue
2276 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
2277 is logged, to avoid truncation. Patch from John Horne.
2279 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
2280 Patch from Jakob Hirsch.
2282 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
2283 SQL string expansion failure details.
2284 Patch from Andrey Oktyabrski.
2286 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
2287 Patch from Simon Arlott.
2289 PP/09 New lookups functionality failed to compile on old gcc which rejects
2290 extern declarations in function scope.
2291 Patch from Oliver Fleischmann
2293 PP/10 Use sig_atomic_t for flags set from signal handlers.
2294 Check getgroups() return and improve debugging.
2295 Fixed developed for diagnosis in bug 927 (which turned out to be
2298 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
2299 Patch from Mark Zealey.
2301 PP/12 Bugzilla 1056: Improved spamd server selection.
2302 Patch from Mark Zealey.
2304 PP/13 Bugzilla 1086: Deal with maildir quota file races.
2305 Based on patch from Heiko Schlittermann.
2307 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
2308 Patch from Uwe Doering, sign-off by Michael Haardt.
2310 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
2311 variable declaration deep within a block. Bug and patch from
2314 PP/15 lookups-Makefile IRIX compatibility coercion.
2316 PP/16 Make DISABLE_DKIM build knob functional.
2318 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
2319 Patch by Simon Arlott
2321 TF/03 Fix valgrind.h portability to C89 compilers that do not support
2322 variable argument macros. Our copy now differs from upstream.
2328 TF/01 Failure to get a lock on a hints database can have serious
2329 consequences so log it to the panic log.
2331 TF/02 Log LMTP confirmation messages in the same way as SMTP,
2332 controlled using the smtp_confirmation log selector.
2334 TF/03 Include the error message when we fail to unlink a spool file.
2336 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
2337 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
2338 for maintaining out-of-tree patches for some time.
2340 PP/01 Bugzilla 139: Documentation and portability issues.
2341 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
2342 Handle per-OS dynamic-module compilation flags.
2344 PP/02 Let /dev/null have normal permissions.
2345 The 4.73 fixes were a little too stringent and complained about the
2346 permissions on /dev/null. Exempt it from some checks.
2347 Reported by Andreas M. Kirchwitz.
2349 PP/03 Report version information for many libraries, including
2350 Exim version information for dynamically loaded libraries. Created
2351 version.h, now support a version extension string for distributors
2352 who patch heavily. Dynamic module ABI change.
2354 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
2355 privilege escalation vulnerability whereby the Exim run-time user
2356 can cause root to append content of the attacker's choosing to
2359 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
2362 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
2363 If dropping privileges for untrusted macros, we disabled normal logging
2364 on the basis that it would fail; for the Exim run-time user, this is not
2365 the case, and it resulted in successful deliveries going unlogged.
2366 Fixed. Reported by Andreas Metzler.
2372 PP/01 Date: & Message-Id: revert to normally being appended to a message,
2373 only prepend for the Resent-* case. Fixes regression introduced in
2374 Exim 4.70 by NM/22 for Bugzilla 607.
2376 PP/02 Include check_rfc2047_length in configure.default because we're seeing
2377 increasing numbers of administrators be bitten by this.
2379 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
2381 PP/03 Bugzilla 994: added openssl_options main configuration option.
2383 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
2385 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
2387 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
2389 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
2390 without a peer certificate, leading to a segfault because of an
2391 assumption that peers always have certificates. Be a little more
2392 paranoid. Problem reported by Martin Tscholak.
2394 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
2395 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
2396 NB: ClamAV planning to remove STREAM in "middle of 2010".
2397 CL also introduces -bmalware, various -d+acl logging additions and
2398 more caution in buffer sizes.
2400 PP/09 Implemented reverse_ip expansion operator.
2402 PP/10 Bugzilla 937: provide a "debug" ACL control.
2404 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
2406 PP/12 Bugzilla 973: Implement --version.
2408 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
2410 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
2412 PP/15 Bugzilla 816: support multiple condition rules on Routers.
2414 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
2415 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
2416 ignore trailing whitespace.
2418 JJ/02 prevent non-panic DKIM error from being sent to paniclog
2420 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
2423 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
2424 Notification from Dr Andrew Aitchison.
2426 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
2427 ExtendedDetectionInfo response format.
2428 Notification from John Horne.
2430 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
2433 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
2434 XSL and documented dependency on system catalogs, with examples of how
2437 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
2440 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
2441 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
2442 configuration file which is writeable by the Exim user or group.
2444 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
2445 of configuration files to cover files specified with the -C option if
2446 they are going to be used with root privileges, not just the default
2449 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
2450 option (effectively making it always true).
2452 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
2453 files to be used while preserving root privileges.
2455 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
2456 that rogue child processes cannot use them.
2458 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
2459 run-time user, instead of root.
2461 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
2462 Exim run-time user without dropping privileges.
2464 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
2465 result string, instead of calling string_vformat() twice with the same
2468 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
2469 for other users. Others should always drop root privileges if they use
2470 -C on the command line, even for a whitelisted configure file.
2472 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
2474 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2480 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2481 $data_path, and $header_path variables; fixed documentation bugs and
2484 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2485 exipick to access non-standard spools, including the "frozen" queue
2488 NM/01 Bugzilla 965: Support mysql stored procedures.
2489 Patch from Alain Williams
2491 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2493 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2494 Patch from Andreas Metzler
2496 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2497 Patch from Kirill Miazine
2499 NM/05 Bugzilla 671: Added umask to procmail example.
2501 JJ/03 installed exipick 20100323.0, fixing doc bug
2503 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2504 directory. Notification and patch from Dan Rosenberg.
2506 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2508 TK/02 Improve log output when DKIM signing operation fails.
2510 MH/01 Treat the transport option dkim_domain as a colon separated
2511 list, not as a single string, and sign the message with each element,
2512 omitting multiple occurences of the same signer.
2514 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2515 Bugzilla 985, 986. Patch by Simon Arlott
2517 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2518 Patch by Simon Arlott
2520 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2521 MBX locking. Notification from Dan Rosenberg.
2527 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2529 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2531 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2533 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2535 NM/04 Bugzilla 915: Flag broken perl installation during build.
2541 TK/01 Added patch by Johannes Berg that expands the main option
2542 "spamd_address" if it starts with a dollar sign.
2544 TK/02 Write list of recipients to X-Envelope-Sender header when building
2545 the mbox-format spool file for content scanning (suggested by Jakob
2548 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2549 (http://www.dcc-servers.net/) support via dccifd. Activated by
2550 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2552 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2553 by Mark Daniel Reidel <mr@df.eu>.
2555 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2556 When building exim an external PCRE library is now needed -
2557 PCRE is a system library on the majority of modern systems.
2558 See entry on PCRE_LIBS in EDITME file.
2560 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2561 conversation. Added nologin parameter to request.
2562 Patch contributed by Kirill Miazine.
2564 TF/01 Do not log submission mode rewrites if they do not change the address.
2566 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2568 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2569 log files in place. Contributed by Roberto Lima.
2571 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2573 TF/03 Bugzilla 615: When checking the local_parts router precondition
2574 after a local_part_suffix or local_part_prefix option, Exim now
2575 does not use the address's named list lookup cache, since this
2576 contains cached lookups for the whole local part.
2578 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2579 Robert Millan. Documentation is in experimental-spec.txt.
2581 TF/04 Bugzilla 668: Fix parallel build (make -j).
2583 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2585 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2586 Patch provided by Jan Srzednicki.
2588 TF/05 Leading white space used to be stripped from $spam_report which
2589 wrecked the formatting. Now it is preserved.
2591 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2592 that they are available at delivery time.
2594 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2596 TF/08 TLS error reporting now respects the incoming_interface and
2597 incoming_port log selectors.
2599 TF/09 Produce a more useful error message if an SMTP transport's hosts
2600 setting expands to an empty string.
2602 NM/06 Bugzilla 744: EXPN did not work under TLS.
2603 Patch provided by Phil Pennock.
2605 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2606 Patch provided by Richard Godbee.
2608 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2609 acl_smtp_notquit, added index entry.
2611 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2612 Patch provided by Eugene Bujak.
2614 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2615 accept(). Patch provided by Maxim Dounin.
2617 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2618 Patch provided by Phil Pennock.
2620 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2622 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2623 Patch provided by Brad "anomie" Jorsch.
2625 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2626 Patch provided by Dean Brooks.
2628 TK/05 Add native DKIM support (does not depend on external libraries).
2630 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2631 Patch provided by Graeme Fowler.
2633 NM/16 Bugzilla 851: Documentation example syntax fix.
2635 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2637 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2640 NM/19 Bugzilla 745: TLS version reporting.
2641 Patch provided by Phil Pennock.
2643 NM/20 Bugzilla 167: bool: condition support.
2644 Patch provided by Phil Pennock.
2646 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2647 clients. Patch provided by Phil Pennock.
2649 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2650 Patch provided by Brad "anomie" Jorsch.
2652 NM/23 Bugzilla 687: Fix misparses in eximstats.
2653 Patch provided by Heiko Schlittermann.
2655 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2656 Patch provided by Heiko Schlittermann.
2658 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2659 plus update to original patch.
2661 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2663 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2664 Patch provided by David Brownlee.
2666 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2668 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2670 NM/30 Bugzilla 888: TLS documentation bugfixes.
2672 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2674 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2675 Unlike the original bugzilla I have changed all shell scripts in src tree.
2677 NM/33 Bugzilla 898: Transport filter timeout fix.
2678 Patch by Todd Rinaldo.
2680 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2681 Patch by Serge Demonchaux.
2683 NM/35 Bugzilla 39: Base64 decode bug fixes.
2684 Patch by Jakob Hirsch.
2686 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2688 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2690 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2692 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2698 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2699 ALT-N's libdkim that I have put here:
2700 http://duncanthrax.net/exim-experimental/
2702 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2703 were called 'true' and it seems that C99 defines that as a reserved
2704 keyword to be used with 'bool' variable types. That means you could
2705 not include C99-style headers which use bools without triggering
2706 build errors in sieve.c.
2708 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2709 as mailq or other aliases. Changed the --help handling significantly
2710 to do whats expected. exim_usage() emits usage/help information.
2712 SC/01 Added the -bylocaldomain option to eximstats.
2714 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2716 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2718 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2724 PH/01 Another patch from the Sieve maintainer.
2726 PH/02 When an IPv6 address is converted to a string for single-key lookup
2727 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2728 dots are used instead of colons so that keys in lsearch files need not
2729 contain colons. This was done some time before quoting was made available
2730 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2731 (notated using the quote facility) so as to distinguish them from IPv4
2732 keys. This meant that lookups for IP addresses in host lists did not work
2733 for iplsearch lookups.
2735 This has been fixed by arranging for IPv6 addresses to be expressed with
2736 colons if the lookup type is iplsearch. This is not incompatible, because
2737 previously such lookups could never work.
2739 The situation is now rather anomalous, since one *can* have colons in
2740 ordinary lsearch keys. However, making the change in all cases is
2741 incompatible and would probably break a number of configurations.
2743 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2746 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2747 conversion specification without a maximum field width, thereby enabling
2748 a rogue spamd server to cause a buffer overflow. While nobody in their
2749 right mind would setup Exim to query an untrusted spamd server, an
2750 attacker that gains access to a server running spamd could potentially
2751 exploit this vulnerability to run arbitrary code as the Exim user.
2753 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2754 $primary_hostname instead of what libspf2 thinks the hosts name is.
2756 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2757 a directory entry by the name of the lookup key. Previously, if a
2758 symlink pointed to a non-existing file or a file in a directory that
2759 Exim lacked permissions to read, a lookup for a key matching that
2760 symlink would fail. Now it is enough that a matching directory entry
2761 exists, symlink or not. (Bugzilla 503.)
2763 PH/03 The body_linecount and body_zerocount variables are now exported in the
2766 PH/04 Added the $dnslist_matched variable.
2768 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2769 This means they are set thereafter only if the connection becomes
2772 PH/06 Added the client_condition to authenticators so that some can be skipped
2773 by clients under certain conditions.
2775 PH/07 The error message for a badly-placed control=no_multiline_responses left
2776 "_responses" off the end of the name.
2778 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2780 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2781 (without spaces) instead of just copying the configuration text.
2783 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2785 PH/11 Added $max_received_linelength.
2787 PH/12 Added +ignore_defer and +include_defer to host lists.
2789 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2790 way in which PCRE > 7.0 is built.
2792 PH/14 Implemented queue_only_load_latch.
2794 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2795 MAIL command. The effect was to mangle the value on 64-bit systems.
2797 PH/16 Another patch from the Sieve maintainer.
2799 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2801 PH/18 If a system quota error occurred while trying to create the file for
2802 a maildir delivery, the message "Mailbox is full" was not appended to the
2803 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2804 only to a quota excession during the actual writing of the file.
2806 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2807 characters?) which causes problems in log lines. The DN values are now
2808 passed through string_printing() before being added to log lines.
2810 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2811 and InterBase are left for another time.)
2813 PH/21 Added message_body_newlines option.
2815 PH/22 Guard against possible overflow in moan_check_errorcopy().
2817 PH/23 POSIX allows open() to be a macro; guard against that.
2819 PH/24 If the recipient of an error message contained an @ in the local part
2820 (suitably quoted, of course), incorrect values were put in $domain and
2821 $local_part during the evaluation of errors_copy.
2827 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2828 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2831 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2832 issue a MAIL command.
2834 PH/02 In an ACL statement such as
2836 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2838 if a client was not listed at all, or was listed with a value other than
2839 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2840 the condition was not true (as it should be), so access was not denied.
2841 The bug was that the ! inversion was incorrectly passed on to the second
2842 item. This has been fixed.
2844 PH/03 Added additional dnslists conditions == and =& which are different from
2845 = and & when the dns lookup returns more than one IP address.
2847 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2848 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2850 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2851 FSYNC, which compiles an option called disable_fsync that allows for
2852 bypassing fsync(). The documentation is heavily laced with warnings.
2854 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2856 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2857 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2858 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2859 including adding "make clean"; (3) Added -fPIC when compiling the test
2860 dynamically loaded module, to get rid of a warning.
2862 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2863 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2864 = 0s. The bug is otherwise harmless.
2866 PH/07 There was a bug in the dovecot authenticator such that the value of
2867 $auth1 could be overwritten, and so not correctly preserved, after a
2868 successful authentication. This usually meant that the value preserved by
2869 the server_setid option was incorrect.
2871 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2873 PH/09 Installed PCRE release 7.0.
2875 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2876 run for batched SMTP input. It is now run at the start of every message
2877 in the batch. While fixing this I discovered that the process information
2878 (output by running exiwhat) was not always getting set for -bs and -bS
2879 input. This is fixed, and it now also says "batched" for BSMTP.
2881 PH/11 Added control=no_pipelining.
2883 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2884 patch, slightly modified), and move the expansion of helo_data till after
2885 the connection is made in the smtp transport (so it can use these
2888 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2890 PH/14 Added log_selector = +pid.
2892 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2894 PH/16 Add ${if forany and ${if forall.
2896 PH/17 Added dsn_from option to vary the From: line in DSNs.
2898 PH/18 Flush SMTP output before performing a callout, unless control =
2899 no_callout_flush is set.
2901 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2902 was true (the default) a successful delivery failed to delete the retry
2903 item, thus causing premature timeout of the address. The bug is now
2906 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2908 PH/21 Long custom messages for fakedefer and fakereject are now split up
2909 into multiline responses in the same way that messages for "deny" and
2910 other ACL rejections are.
2912 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2913 with slight modification.
2915 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2916 draft, changing the syntax and factoring some duplicate code".
2918 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2919 for deliveries of the second and subsequent messages over the same SMTP
2922 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2923 ${reduce, with only minor "tidies".
2925 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2927 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2928 expansion side effects.
2930 PH/27 When a message times out after an over-quota error from an Exim-imposed
2931 quota, the bounce message says "mailbox is full". This message was not
2932 being given when it was a system quota that was exceeded. It now should
2935 MH/03 Made $recipients available in local_scan(). local_scan() already has
2936 better access to the recipient list through recipients_list[], but
2937 $recipients can be useful in postmaster-provided expansion strings.
2939 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2940 in the case of a MAIL command with additional options following the
2941 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2942 were accidentally chopped off.
2944 PH/29 SMTP synchronization checks are implemented when a command is read -
2945 there is a check that no more input is waiting when there shouldn't be
2946 any. However, for some commands, a delay in an ACL can mean that it is
2947 some time before the response is written. In this time, more input might
2948 arrive, invalidly. So now there are extra checks after an ACL has run for
2949 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2950 pipelining has not been advertised.
2952 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2954 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2955 returned in the SMTP response when smtp_return_error_details was false.
2956 This has been fixed.
2958 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2959 instead of the C I/O that was originally supplied, because problems were
2960 reported on Solaris.
2962 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2963 Exim which did not show up earlier: it was assuming that a call to
2964 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2965 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2966 was a macro that became an assignment, so it seemed to work. This has
2967 changed to a proper function call with a void return, hence the compile
2968 error. Exim's code has been fixed.
2970 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2973 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2975 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2977 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2978 because it turns out to be too big or there is a timeout) there is no
2979 "Completed" line in the log. When some messages of this type were
2980 selected by exigrep, they were listed as "not completed". Others were
2981 picked up by some special patterns. I have improved the selection
2982 criteria to be more general.
2984 PH/38 The host_find_failed option in the manualroute router can now be set
2985 to "ignore", to completely ignore a host whose IP address cannot be
2986 found. If all hosts are ignored, the behaviour is controlled by the new
2987 host_all_ignored option.
2989 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2990 homing or because of multiple MX records with /mx) generated more than
2991 one IP address, and the following item turned out to be the local host,
2992 all the secondary addresses of the first item were incorrectly removed
2993 from the list, along with the local host and any following hosts (which
2994 is what is supposed to happen).
2996 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2997 whoever called Exim into the -H file. In the case of the daemon it was
2998 behaving confusingly. When first started, it used values for whoever
2999 started the daemon, but after a SIGHUP it used the Exim user (because it
3000 calls itself on a restart). I have changed the code so that it now always
3003 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
3004 message are rejected with the same error (e.g. no authentication or bad
3005 sender address), and a DATA command is nevertheless sent (as can happen
3006 with PIPELINING or a stupid MUA), the error message that was given to the
3007 RCPT commands is included in the rejection of the DATA command. This is
3008 intended to be helpful for MUAs that show only the final error to their
3011 PH/42 Another patch from the Sieve maintainer.
3013 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
3014 Eximstats - Fixed some broken HTML links and added missing column headers
3016 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
3017 columns for Rejects, Temp Rejects, Ham, and Spam rows.
3019 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
3021 PH/43 Yet another patch from the Sieve maintainer.
3023 PH/44 I found a way to check for a TCP/IP connection going away before sending
3024 the response to the final '.' that terminates a message, but only in the
3025 case where the client has not sent further data following the '.'
3026 (unfortunately, this is allowed). However, in many cases there won't be
3027 any further data because there won't be any more messages to send. A call
3028 to select() can be used: if it shows that the input is "ready", there is
3029 either input waiting, or the socket has been closed. An attempt to read
3030 the next input character can distinguish the two cases. Previously, Exim
3031 would have sent an OK response which the client would never have see.
3032 This could lead to message repetition. This fix should cure that, at
3033 least in a lot of common cases.
3035 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
3036 advertised in response to EHLO.
3042 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
3043 fixed by 4.65/MH/01 (is this a record?) are fixed:
3045 (i) An empty string was always treated as zero by the numeric comparison
3046 operators. This behaviour has been restored.
3048 (ii) It is documented that the numeric comparison operators always treat
3049 their arguments as decimal numbers. This was broken in that numbers
3050 starting with 0 were being interpreted as octal.
3052 While fixing these problems I realized that there was another issue that
3053 hadn't been noticed. Values of message_size_limit (both the global option
3054 and the transport option) were treated as octal if they started with 0.
3055 The documentation was vague. These values are now always treated as
3056 decimal, and I will make that clear in the documentation.
3062 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
3063 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
3066 MH/01 Don't check that the operands of numeric comparison operators are
3067 integers when their expansion is in "skipping" mode (fixes bug
3068 introduced by 4.64-PH/07).
3070 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
3071 child addresses, Exim now panics and dies. Previously, because the count
3072 is held in a short int, deliveries were likely to be lost. As such a
3073 large number of recipients for a single message is ridiculous
3074 (performance will be very, very poor), I have chosen to impose a limit
3075 rather than extend the field.
3081 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
3082 leftover -K file (the existence of which was triggered by #402).
3083 While we were at it, introduced process PID as part of the -K
3084 filename. This should rule out race conditions when creating
3087 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
3088 processing considerably. Previous code took too long for large mails,
3089 triggering a timeout which in turn triggers #401.
3091 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
3092 in the DK code in transports.c. sendfile() is not really portable,
3093 hence the _LINUX specificness.
3095 TF/01 In the add_headers option to the mail command in an Exim filter,
3096 there was a bug that Exim would claim a syntax error in any
3097 header after the first one which had an odd number of characters
3100 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
3101 callout verification, Exim cached a "reject" for the entire domain. This
3102 is correct for most verifications, but it is not correct for a recipient
3103 verification with use_sender or use_postmaster set, because in that case
3104 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
3105 case of MAIL FROM:<> rejection from other early rejections (e.g.
3106 rejection of HELO). When verifying a recipient using a non-null MAIL
3107 address, the cache is ignored if it shows MAIL FROM:<> rejection.
3108 Whatever the result of the callout, the value of the domain cache is
3109 left unchanged (for any other kind of callout, getting as far as trying
3110 RCPT means that the domain itself is ok).
3112 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
3115 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
3116 manifest itself as EPIPE rather than ECONNECT. When tidying away a
3117 session, the daemon ignores ECONNECT errors and logs others; it now
3118 ignores EPIPE as well.
3120 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
3121 (quoted-printable decoding).
3123 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
3124 later the small subsequent patch to fix an introduced bug.
3126 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
3128 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
3130 PH/08 An error is now given if message_size_limit is specified negative.
3132 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
3133 to be given (somewhat) arbitrary names.
3135 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
3138 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
3139 miscellaneous code fixes
3141 PH/10 Added the log_reject_target ACL modifier to specify where to log
3144 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
3145 hostname. This is wrong, because it relates to the incoming message (and
3146 probably the interface on which it is arriving) and not to the outgoing
3147 callout (which could be using a different interface). This has been
3148 changed to use the value of the helo_data option from the smtp transport
3149 instead - this is what is used when a message is actually being sent. If
3150 there is no remote transport (possible with a router that sets up host
3151 addresses), $smtp_active_hostname is used.
3153 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
3154 tweaks were necessary in order to get it to work (see also 21 below):
3155 (a) The code assumed that strncpy() returns a negative number on buffer
3156 overflow, which isn't the case. Replaced with Exim's string_format()
3158 (b) There were several signed/unsigned issues. I just did the minimum
3159 hacking in of casts. There is scope for a larger refactoring.
3160 (c) The code used strcasecmp() which is not a standard C function.
3161 Replaced with Exim's strcmpic() function.
3162 (d) The code set only $1; it now sets $auth1 as well.
3163 (e) A simple test gave the error "authentication client didn't specify
3164 service in request". It would seem that Dovecot has changed its
3165 interface. Fortunately there's a specification; I followed it and
3166 changed what the client sends and it appears to be working now.
3168 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
3171 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
3172 address A is aliased to B and C, where B exists and C does not. Without
3173 -v the output is "A verified" because verification stops after a
3174 successful redirection if more than one address is generated. However,
3175 with -v the child addresses are also verified. Exim was outputting "A
3176 failed to verify" and then showing the successful verification for C,
3177 with its parentage. It now outputs "B failed to verify", showing B's
3178 parentage before showing the successful verification of C.
3180 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
3181 look up a TXT record in a specific list after matching in a combined
3184 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
3185 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
3186 they consult the DNS. I had assumed they would set it the way they
3187 wanted; and indeed my experiments on Linux seem to show that in some
3188 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
3189 To be on the safe side, however, I have now made the interface to
3190 host_find_byname() similar to host_find_bydns(), with an argument
3191 containing the DNS resolver options. The host_find_byname() function now
3192 sets these options at its start, just as host_find_bydns() does. The smtp
3193 transport options dns_qualify_single and dns_search_parents are passed to
3194 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
3195 of host_find_byname() use the default settings of RES_DEFNAMES
3196 (qualify_single) but not RES_DNSRCH (search_parents).
3198 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
3199 spool_read_header() do less string testing, by means of a preliminary
3200 switch on the second character of optional "-foo" lines. (This is
3201 overdue, caused by the large number of possibilities that now exist.
3202 Originally there were few.) While I was there, I also converted the
3203 str(n)cmp tests so they don't re-test the leading "-" and the first
3204 character, in the hope this might squeeze out yet more improvement.
3206 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
3207 flag allowing group syntax was set by the header_syntax check but not
3208 turned off, possible causing trouble later; (2) The flag was not being
3209 set at all for the header_verify test, causing "group"-style headers to
3210 be rejected. I have now set it in this case, and also caused header_
3211 verify to ignore an empty address taken from a group. While doing this, I
3212 came across some other cases where the code for allowing group syntax
3213 while scanning a header line wasn't quite right (mostly, not resetting
3214 the flag correctly in the right place). These bugs could have caused
3215 trouble for malformed header lines. I hope it is now all correct.
3217 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
3218 with the "reply" argument non-NULL. The code, however (which originally
3219 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
3220 but it didn't always do it. This confused somebody who was copying the
3221 code for some other use. I have removed all the tests.
3223 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
3224 feature that was used to support insecure browsers during the U.S. crypto
3225 embargo. It requires special client support, and Exim is probably the
3226 only MTA that supported it -- and would never use it because real RSA is
3227 always available. This code has been removed, because it had the bad
3228 effect of slowing Exim down by computing (never used) parameters for the
3229 RSA_EXPORT functionality.
3231 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
3232 authenticator to fail if there's a tab character in the incoming data
3233 (there should never be unless someone is messing about, as it's supposed
3234 to be base64-encoded). Also added, on Timo's advice, the "secured" option
3235 if the connection is using TLS or if the remote IP is the same as the
3236 local IP, and the "valid-client-cert option" if a client certificate has
3239 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
3240 authenticators. This can be used for authorization after authentication
3241 succeeds. (In the case of plaintext, it servers for both authentication
3244 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
3245 if any retry times were supplied.
3247 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
3248 connection, where there is no client IP address to check. In this
3249 situation, the verify now always succeeds.
3251 PH/25 Applied John Jetmore's -Mset patch.
3253 PH/26 Added -bem to be like -Mset, but loading a message from a file.
3255 PH/27 In a string expansion for a processed (not raw) header when multiple
3256 headers of the same name were present, leading whitespace was being
3257 removed from all of them, but trailing whitespace was being removed only
3258 from the last one. Now trailing whitespace is removed from each header
3259 before concatenation. Completely empty headers in a concatenation (as
3260 before) are ignored.
3262 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
3263 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
3265 PH/29 [Removed. This was a change that I later backed out, and forgot to
3266 correct the ChangeLog entry (that I had efficiently created) before
3267 committing the later change.]
3269 PH/30 Exim was sometimes attempting to deliver messages that had suffered
3270 address errors (4xx response to RCPT) over the same connection as other
3271 messages routed to the same hosts. Such deliveries are always "forced",
3272 so retry times are not inspected. This resulted in far too many retries
3273 for the affected addresses. The effect occurred only when there were more
3274 hosts than the hosts_max_try setting in the smtp transport when it had
3275 the 4xx errors. Those hosts that it had tried were not added to the list
3276 of hosts for which the message was waiting, so if all were tried, there
3277 was no problem. Two fixes have been applied:
3279 (i) If there are any address or message errors in an SMTP delivery, none
3280 of the hosts (tried or untried) are now added to the list of hosts
3281 for which the message is waiting, so the message should not be a
3282 candidate for sending over the same connection that was used for a
3283 successful delivery of some other message. This seems entirely
3284 reasonable: after all the message is NOT "waiting for some host".
3285 This is so "obvious" that I'm not sure why it wasn't done
3286 previously. Hope I haven't missed anything, but it can't do any
3287 harm, as the worst effect is to miss an optimization.
3289 (ii) If, despite (i), such a delivery is accidentally attempted, the
3290 routing retry time is respected, so at least it doesn't keep
3291 hammering the server.
3293 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
3294 in ${readsocket because some servers need this prod.
3296 PH/32 Added some extra debug output when updating a wait-xxx database.
3298 PH/33 The hint "could be header name not terminated by colon", which has been
3299 given for certain expansion errors for a long time, was not being given
3300 for the ${if def:h_colon_omitted{... case.
3302 PH/34 The spec says: "With one important exception, whenever a domain list is
3303 being scanned, $domain contains the subject domain." There was at least
3304 one case where this was not true.
3306 PH/35 The error "getsockname() failed: connection reset by peer" was being
3307 written to the panic log as well as the main log, but it isn't really
3308 panic-worthy as it just means the connection died rather early on. I have
3309 removed the panic log writing for the ECONNRESET error when getsockname()
3312 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
3313 runs only) independently of the message's sender address. This meant
3314 that, if the 4xx error was in fact related to the sender, a different
3315 message to the same recipient with a different sender could confuse
3316 things. In particular, this can happen when sending to a greylisting
3317 server, but other circumstances could also provoke similar problems.
3318 I have changed the default so that the retry time for these errors is now
3319 based a combination of the sender and recipient addresses. This change
3320 can be overridden by setting address_retry_include_sender=false in the
3323 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
3324 remote server are returned as part of bounce messages. This was not
3325 happening for LMTP over a pipe (the lmtp transport), but now it is the
3326 same for both kinds of LMTP.
3328 PH/38 Despite being documented as not happening, Exim was rewriting addresses
3329 in header lines that were in fact CNAMEs. This is no longer the case.
3331 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
3332 and queue runs started by the daemon processed all messages. This has
3333 been fixed so that -R and -S can now usefully be given with -q<time>.
3335 PH/40 Import PCRE release 6.7 (fixes some bugs).
3337 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
3339 PH/42 Give an error if -q is specified more than once.
3341 PH/43 Renamed the variables $interface_address and $interface_port as
3342 $received_ip_address and $received_port, to make it clear that these
3343 values apply to message reception, and not to the outgoing interface when
3344 a message is delivered. (The old names remain recognized, of course.)
3346 PH/44 There was no timeout on the connect() call when using a Unix domain
3347 socket in the ${readsocket expansion. There now is.
3349 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
3350 be meaningful with "accept".
3352 SC/01 Eximstats V1.43
3353 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
3355 SC/02 Eximstats V1.44
3356 Use a glob alias rather than an array ref in the generated
3357 parser. This improves both readability and performance.
3359 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
3360 Collect SpamAssassin and rejection statistics.
3361 Don't display local sender or destination tables unless
3362 there is data to show.
3363 Added average volumes into the top table text output.
3365 SC/04 Eximstats V1.46
3366 Collect data on the number of addresses (recipients)
3367 as well as the number of messages.
3369 SC/05 Eximstats V1.47
3370 Added 'Message too big' to the list of mail rejection
3371 reasons (thanks to Marco Gaiarin).
3373 SC/06 Eximstats V1.48
3374 Mainlog lines which have GMT offsets and are too short to
3375 have a flag are now skipped.
3377 SC/07 Eximstats V1.49 (Alain Williams)
3378 Added the -emptyok flag.
3380 SC/08 Eximstats V1.50
3381 Fixes for obtaining the IP address from reject messages.
3383 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
3384 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
3385 whitespace changes from 4.64-PH/27
3387 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
3390 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
3391 are found, allow negative numbers in numeric criteria)
3393 JJ/06 exipick.20061117.2, added new $message_body_missing variable
3395 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
3396 to match changes made in 4.64-PH/43
3398 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
3400 PH/47 Put in an explicit test for a DNS lookup of an address record where the
3401 "domain" is actually an IP address, and force a failure. This locks out
3402 those revolvers/nameservers that support "A-for-A" lookups, in
3403 contravention of the specifications.
3405 PH/48 When a host name was looked up from an IP address, and the subsequent
3406 forward lookup of the name timed out, the host name was left in
3407 $sender_host_name, contrary to the specification.
3409 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
3410 restricted to single-key lookups, Exim was not diagnosing an error if
3411 * or *@ was used with a query-style lookup.
3413 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
3415 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
3416 long ago, but noone interested enough thought of it. Let's just say that
3417 the "1.1" means that there are some new functions that weren't there at
3418 some point in the past.
3420 PH/51 Error processing for expansion failure of helo_data from an smtp
3421 transport during callout processing was broken.
3423 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
3424 tested/used via the -bh/-bhc/-bs options.
3426 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
3427 bug, fixed in subsequent PCRE releases).
3429 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
3430 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
3432 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
3438 SC/01 Use a glob alias rather than an array ref in eximstats generated
3439 parser. This improves both readability and performance.
3441 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
3442 Don't display local sender or destination tables in eximstats unless
3443 there is data to show.
3444 Added average volumes into the eximstats top table text output.
3446 SC/03 Collect data on the number of addresses (recipients) as well
3447 as the number of messages in eximstats.
3449 TF/01 Correct an error in the documentation for the redirect router. Exim
3450 does (usually) call initgroups() when daemonizing.
3452 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
3453 with consistent privilege compared to when running as a daemon.
3455 TF/03 Note in the spec that $authenticated_id is not set for local
3456 submissions from trusted users.
3458 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
3459 Thanks to Dean Brooks <dean@iglou.com> for the patch.
3461 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
3462 by adding some example configuration directives to the default
3463 configuration file. A little bit of work is required to uncomment the
3464 directives and define how usernames and passwords are checked, but
3465 there is now a framework to start from.
3467 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
3468 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
3469 without this. I don't know how relevant this is to other LDAP libraries.
3471 PH/02 Add the verb name to the "unknown ACL verb" error.
3473 PH/03 Magnus Holmgren's patch for filter_prepend_home.
3475 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
3477 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
3478 directory not expanded when it should be if an expanded home directory
3479 was set for the address (which is overridden by the transport).
3481 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3484 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3485 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3486 because it is too late at that time, and has no effect.
3488 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3489 security issue with \' (bugzilla #107). I could not use the
3490 PQescapeStringConn() function, because it needs a PGconn value as one of
3493 PH/08 When testing addresses using -bt, indicate those final addresses that
3494 are duplicates that would not cause an additional delivery. At least one
3495 person was confused, thinking that -bt output corresponded to deliveries.
3496 (Suppressing duplicates isn't a good idea as you lose the information
3497 about possibly different redirections that led to the duplicates.)
3499 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3500 systems where poll() doesn't work, in particular OS X.
3502 PH/10 Added more information to debugging output for retry time not reached.
3504 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3505 operations in malware.c.
3507 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3510 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3511 syslog_duplication set false, log lines that would normally be written
3512 both the the main log and to the reject log were not written to syslog at
3515 PH/14 In the default configuration, change the use of "message" in ACL warn
3516 statements to "add_header".
3518 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3519 not followed by a command (e.g. "seen endif").
3521 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3522 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3525 PH/17 Added extra conditions to the default value of delay_warning_condition
3529 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3530 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3531 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3534 The Auto-Submitted: and various List- headers are standardised, whereas I
3535 don't think Precedence: ever was.
3537 PH/18 Refactored debugging code in route_finduser() to show more information,
3538 in particular, the error code if getpwnam() issues one.
3540 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3541 This is apparently needed in addition to the PH/07 change above to avoid
3542 any possible encoding problems.
3544 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3545 but not after initializing Perl.
3547 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3548 output them only if debugging. By default they are written stderr,
3549 apparently, which is not desirable.
3551 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3554 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3557 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3559 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3560 authenticated or an ident call has been made. Suppress the default
3561 values for $authenticated_id and $authenticated_sender (but permit -oMai
3562 and -oMas) when testing with -bh.
3564 PH/24 Re-jigged the order of the tests in the default configuration so that the
3565 tests for valid domains and recipients precede the DNS black list and CSA
3566 tests, on the grounds that those ones are more expensive.
3568 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3569 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3570 This bug exists in every version of Exim that I still have, right back to
3573 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3574 However, an attempt to turn on case-sensitivity in a regex key by
3575 including (?-i) didn't work because the subject string was already
3576 lowercased, and the effects were non-intuitive. It turns out that a
3577 one-line patch can be used to allow (?-i) to work as expected.
3583 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3584 other effects) broke the use of negated acl sub-conditions.
3586 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3589 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3590 "Deny" causes Exim to reject the incoming connection with a 554 error.
3591 Unfortunately, if there is a major crisis, such as a disk failure,
3592 tcp-wrappers gives "deny", whereas what one would like would be some
3593 kind of temporary error. A kludge has been added to help with this.
3594 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3595 554 error is used if errno is still zero or contains ENOENT (which occurs
3596 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3599 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3601 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3602 errors. Otherwise a message that provokes a temporary error (when other
3603 messages do not) can cause a whole host to time out.
3605 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3606 addresses were routed directly to files or pipes from a redirect router.
3607 File deliveries just didn't batch; pipe deliveries might have suffered
3610 PH/06 A failure to get a lock for a hints database would erroneously always say
3611 "Failed to get write lock", even when it was really a read lock.
3613 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3614 of 0600. This has been changed to use the value of the lockfile_mode
3615 option (which defaults to 0600).
3617 PH/08 Applied small patch from the Sieve maintainer.
3619 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3620 folder from quota calculations, a direct delivery into this folder messed
3621 up the contents of the maildirsize file. This was because the regex was
3622 used only to exclude .Trash (or whatever) when the size of the mailbox
3623 was calculated. There was no check that a delivery was happening into an
3624 excluded directory. This bug has been fixed by ignoring all quota
3625 processing for deliveries into excluded directories.
3627 PH/10 Added the maildirfolder_create_regex option to appendfile.
3633 PH/01 The code for finding all the local interface addresses on a FreeBSD
3634 system running IPv6 was broken. This may well have applied to all BSD
3635 systems, as well as to others that have similar system calls. The broken
3636 code found IPv4 interfaces correctly, but gave incorrect values for the
3637 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3638 that it would not match correctly against @[] and not recognize the IPv6
3641 PH/02 The ipliteral router was not recognizing addresses of the form user@
3642 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3644 PH/03 Added disable_ipv6.
3646 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3647 decoded form, because it is most often used to construct To: headers
3648 lines in autoreplies, and the decoded form may well be syntactically
3649 invalid. However, $reply_address has leading white space removed, and all
3650 newlines turned into spaces so that the autoreply transport does not
3653 PH/05 If group was specified without a user on a router, and no group or user
3654 was specified on a transport, the group from the router was ignored.
3656 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3657 for visible compile-time settings that can be used to change these
3658 numbers, for those that want even more. Backwards compatibility with old
3659 spool files has been maintained. However, going back to a previous Exim
3660 release will lost any variables that are in spool files.
3662 PH/07 Two small changes when running in the test harness: increase delay when
3663 passing a TCP/IP connection to a new process, in case the original
3664 process has to generate a bounce, and remove special handling of
3665 127.0.0.2 (sic), which is no longer necessary.
3667 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3668 be the same on different OS.
3670 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3673 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3674 whether --show-vars was specified or not
3676 JJ/02 exipick: Added support for new ACL variable spool format introduced
3679 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3680 syntactically invalid From: or Reply-to: line, and a filter used this to
3681 generate an autoreply, and therefore failed to obtain an address for the
3682 autoreply, Exim could try to deliver to a non-existent relative file
3683 name, causing unrelated and misleading errors. What now happens is that
3684 it logs this as a hard delivery error, but does not attempt to create a
3687 PH/11 The exinext utility has a -C option for testing purposes, but although
3688 the given file was scanned by exinext itself; it wasn't being passed on
3689 when Exim was called.
3691 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3692 an end-of-file indication when reading a command response.
3694 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3695 compiled. In many other places in Exim, IPv6 addresses are always
3696 recognized, so I have changed this. It also means that IPv4 domain
3697 literals of the form [IPV4:n.n.n.n] are now always recognized.
3699 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3700 used if the router is not running as root, for example, when verifying at
3701 ACL time, or when using -bh. The debugging output from this situation was
3702 non-existent - all you got was a failure to exec. I have made two
3705 (a) Failures to set uid/gid, the current directory, or a process leader
3706 in a subprocess such as that created by queryprogram now generate
3707 suitable debugging output when -d is set.
3709 (b) The queryprogram router detects when it is not running as root,
3710 outputs suitable debugging information if -d is set, and then runs
3711 the subprocess without attempting to change uid/gid.
3713 PH/15 Minor change to Makefile for building test_host (undocumented testing
3716 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3717 additional section of a DNS packet that returns MX or SRV records.
3718 Instead, it always explicitly searches for A/AAAA records. This avoids
3719 major problems that occur when a DNS server includes only records of one
3720 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3721 fixed another bug: if SRV records were looked up and the corresponding
3722 address records were *not* found in the additional section, the port
3723 values from the SRV records were lost.
3725 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3726 using the correct key (the original address) when searching the retry
3727 rules in order to find which one to use for generating the retry hint.
3729 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3730 adding the default one. Similarly, if it contains a Reply-To: header, the
3731 errors_reply_to option, if set, is not used.
3733 PH/19 When calculating a retry time, Exim used to measure the "time since
3734 failure" by looking at the "first failed" field in the retry record. Now
3735 it does not use this if it is later than than the arrival time of the
3736 message. Instead it uses the arrival time. This makes for better
3737 behaviour in cases where some deliveries succeed, thus re-setting the
3738 "first failed" field. An example is a quota failure for a huge message
3739 when small messages continue to be delivered. Without this change, the
3740 "time since failure" will always be short, possible causing more frequent
3741 delivery attempts for the huge message than are intended.
3742 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3744 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3745 $1, $2, $3) because the numerical variables can be reset during some
3746 expansion items (e.g. "match"), thereby losing the authentication data.
3748 PH/21 Make -bV show the size of off_t variables so that the test suite can
3749 decide whether to run tests for quotas > 2G.
3751 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3752 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3753 filecount value is greater than 2G or if a quota value is greater than 2G
3754 on a system where the size of off_t is not greater than 4, a panic error
3757 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3758 never match. The debug and -bh output now contains an explicit error
3759 message indicating a malformed IPv4 address or mask.
3761 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3762 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3763 PH/23 above applies.
3765 PH/25 Do not write to syslog when running in the test harness. The only
3766 occasion when this arises is a failure to open the main or panic logs
3767 (for which there is an explicit test).
3769 PH/26 Added the /no_tell option to "control=freeze".
3771 PH/27 If a host name lookup failed very early in a connection, for example, if
3772 the IP address matched host_lookup and the reverse lookup yielded a name
3773 that did not have a forward lookup, an error message of the form "no IP
3774 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3775 could be logged. Now it outputs the IP address instead of "NULL".
3777 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3778 allows the sender and the authenticated sender to be set when
3779 submitting a message from within Exim. Since child_open_exim() is
3780 documented for local_scan(), the new function should be too.
3782 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3783 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3784 results in an empty string is now treated as unset.
3786 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3788 PH/31 Added sender_verify_fail logging option.
3790 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3791 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3792 tidied the source and removed it altogether.
3794 PH/33 When a queue run was abandoned because the load average was too high, a
3795 log line was always written; now it is written only if the queue_run log
3796 selector is set. In addition, the log line for abandonment now contains
3797 information about the queue run such as the pid. This is always present
3798 in "start" and "stop" lines but was omitted from the "abandon" line.
3800 PH/34 Omit spaces between a header name and the colon in the error message that
3801 is given when verify = headers_syntax fails (if there are lots of them,
3802 the message gets confusing).
3804 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3805 names, as there are now some PTR records that contain slashes. This check
3806 is only to protect against broken name servers that fall over on strange
3807 characters, so the fact that it applies to all lookups doesn't matter.
3809 PH/36 Now that the new test suite is complete, we can remove some of the
3810 special code in Exim that was needed for the old test suite. For example,
3811 sorting DNS records because real resolvers return them in an arbitrary
3812 order. The new test suite's fake resolver always returns records in the
3815 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3816 bounces) except when queue_only is set, to avoid logging races between
3817 the different processes.
3819 PH/38 Panic-die if .include specifies a non-absolute path.
3821 PH/39 A tweak to the "H" retry rule from its user.
3823 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3824 a label. They prevented compilation on older perls.
3826 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3827 a warning to be raised on newish perls.
3829 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3830 on queue. Changes to match documented behaviour of showing count of
3831 messages matching specified criteria.
3833 PH/40 Changed the default ident timeout from 30s to 5s.
3835 PH/41 Added support for the use of login_cap features, on those BSD systems
3836 that have them, for controlling the resources used by pipe deliveries.
3838 PH/42 The content-scanning code uses fopen() to create files in which to put
3839 message data. Previously it was not paying any attention to the mode of
3840 the files. Exim runs with umask(0) because the rest of the code creates
3841 files with open(), and sets the required mode explicitly. Thus, these
3842 files were ending up world-writeable. This was not a big issue, because,
3843 being within the spool directory, they were not world-accessible. I have
3844 created a function called modefopen, which takes an additional mode
3845 argument. It sets umask(777), creates the file, chmods it to the required
3846 mode, then resets the umask. All the relevant calls to fopen() in the
3847 content scanning code have been changed to use this function.
3849 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3850 to 24 hours. This avoids potential overflow problems when processing G
3851 and H retry rules. I suspect nobody ever tinkers with this value.
3853 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3855 PH/45 When the plaintext authenticator is running as a client, the server's
3856 challenges are checked to ensure they are valid base64 strings. By
3857 default, the authentication attempt is cancelled if an invalid string is
3858 received. Setting client_ignore_invalid_base64 true ignores these errors.
3859 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3860 they are received. Thus, the responses can be made to depend on the
3861 challenges. If an invalid string is ignored, an empty string is placed in
3864 PH/46 Messages that are created by the autoreply transport now contains a
3865 References: header, in accordance with RFCs 2822 and 3834.
3867 PH/47 Added authenticated_sender_force to the smtp transport.
3869 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3871 PH/49 Installed latest patch from the Sieve maintainer.
3873 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3874 was also set, the appendfile transport was unnecessarily scanning a
3875 directory of message files (e.g. for maildir delivery) to find the count
3876 of files (along with the size), even though it did not need this
3877 information. It now does the scan only if it needs to find either the
3878 size of the count of files.
3880 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3882 PH/52 Two bugs concerned with error handling when the smtp transport is
3885 (i) Exim was not creating retry information for temporary errors given
3886 for individual recipients after the DATA command when the smtp transport
3887 was used in LMTP mode. This meant that they could be retried too
3888 frequently, and not timed out correctly.
3890 (ii) Exim was setting the flag that allows error details to be returned
3891 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3892 recipients that were returned after the DATA command.
3894 PH/53 This is related to PH/52, but is more general: for any failing address,
3895 when detailed error information was permitted to be returned to the
3896 sender, but the error was temporary, then after the final timeout, only
3897 "retry timeout exceeded" was returned. Now it returns the full error as
3898 well as "retry timeout exceeded".
3900 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3901 do this, and (what is worse) MTAs that accept it.
3903 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3904 will now be deprecated.
3906 PH/56 New os.c-cygwin from the Cygwin maintainer.
3908 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3909 formats (previously only available in exim formats via -bpr, -bpru,
3910 and -bpra. Now also available in native and exiqgrep formats)
3912 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3913 with very large, slow to parse queues
3915 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3917 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3919 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3920 responses to authentication challenges, though it was showing the
3921 challenges; (ii) I've removed the CR characters from the debug output for
3924 PH/58 Allow for the insertion of a newline as well as a space when a string
3925 is turned into more than one encoded-word during RFC 2047 encoding. The
3926 Sieve code now uses this.
3928 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3929 data_4xx, lost_connection, tls_required.
3931 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3932 message was being sent as an SMTP response.
3934 PH/61 Add -l and -k options to exicyclog.
3936 PH/62 When verifying, if an address was redirected to one new address, so that
3937 verification continued, and the new address failed or deferred after
3938 having set something in $address_data, the value of $address_data was not
3939 passed back to the ACL. This was different to the case when no
3940 redirection occurred. The value is now passed back in both cases.
3942 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3943 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3944 use login_cap.h, so on its own it isn't the distinguishing feature. The
3945 new name refers directly to the setclassresources() function.
3947 PH/65 Added configuration files for NetBSD3.
3949 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3951 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3952 is preferred over IPv4.
3954 PH/68 The bounce_return_message and bounce_return_body options were not being
3955 honoured for bounces generated during the reception of non-SMTP messages.
3956 In particular, this applied to messages rejected by the ACL. This bug has
3957 been fixed. However, if bounce_return_message is true and bounce_return_
3958 body is false, the headers that are returned for a non-SMTP message
3959 include only those that have been read before the error was detected.
3960 (In the case of an ACL rejection, they have all been read.)
3962 PH/69 The HTML version of the specification is now built in a directory called
3963 spec_html instead of spec.html, because the latter looks like a path with
3964 a MIME-type, and this confuses some software.
3966 PH/70 Catch two compiler warnings in sieve.c.
3968 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3969 function verify_get_ident() calls ip_connect() to connect a socket, but
3970 if the "connect()" function timed out, ip_connect() used to close the
3971 socket. However, verify_get_ident() also closes the socket later, and in
3972 between Exim writes to the log, which may get opened at this point. When
3973 the socket was closed in ip_connect(), the log could get the same file
3974 descriptor number as the socket. This naturally causes chaos. The fix is
3975 not to close the socket in ip_connect(); the socket should be closed by
3976 the function that creates it. There was only one place in the code where
3977 this was missing, in the iplookup router, which I don't think anybody now
3978 uses, but I've fixed it anyway.
3980 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3981 well as to direct DNS lookups. Otherwise the handling of names in host
3982 lists is inconsistent and therefore confusing.
3988 PH/01 Two changes to the default runtime configuration:
3990 (1) Move the checks for relay_from_hosts and authenticated clients from
3991 after to before the (commented out) DNS black list checks.
3993 (2) Add control=submission to the relay_from_hosts and authenticated
3994 clients checks, on the grounds that messages accepted by these
3995 statements are most likely to be submissions.
3997 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3999 (1) Generate an error if the third argument for the ${prvs expansion is
4002 (2) Treat a missing third argument of ${prvscheck as if it were an empty
4005 (3) Reset the variables that are obtained from the first argument of
4006 ${prvscheck and used in the second argument before leaving the code,
4007 because their memory is reclaimed, so using them afterwards may do
4010 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
4011 one (it's much easier than Tom thought :-).
4013 (5) Because of (4), we can now allow for the use of $prvscheck_result
4014 inside the third argument.
4016 PH/03 For some reason, the default setting of PATH when running a command from
4017 a pipe transport was just "/usr/bin". I have changed it to
4020 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
4021 anything to be listed in the output from -bV.
4023 PH/05 When a filter generated an autoreply, the entire To: header line was
4024 quoted in the delivery log line, like this:
4026 => >A.N.Other <ano@some.domain> <original@ddress> ...
4028 This has been changed so that it extracts the operative address. There
4029 may be more than one such address. If so, they are comma-separated, like
4032 => >ano@some.domain,ona@other.domain <original@ddress> ...
4034 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
4035 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
4036 not being looked up in the rDNS to get a host name, Exim was showing the
4037 IP address twice in Received: lines, even though the IP addresses were
4038 identical. For example:
4040 Received: from [1.2.3.4] (helo=[1.2.3.4])
4042 However, if the real host name was known, it was omitting the HELO data
4043 if it matched the actual IP address. This has been tidied up so that it
4044 doesn't show the same IP address twice.
4046 PH/07 When both +timestamp and +memory debugging was on, the value given by
4047 $tod_xxx expansions could be wrong, because the tod_stamp() function was
4048 called by the debug printing, thereby overwriting the timestamp buffer.
4049 Debugging no longer uses the tod_stamp() function when +timestamp is set.
4051 PH/08 When the original message was included in an autoreply transport, it
4052 always said "this is a copy of the message, including all the headers",
4053 even if body_only or headers_only was set. It now gives an appropriate
4056 PH/09 Applied a patch from the Sieve maintainer which:
4058 o fixes some comments
4059 o adds the (disabled) notify extension core
4060 o adds some debug output for the result of if/elsif tests
4061 o points to the current vacation draft in the documentation
4062 and documents the missing references header update
4066 o fixes a bug in processing the envelope test (when testing
4067 multiple envelope elements, the last element determined the
4070 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
4071 Electronic Mail") by including:
4073 Auto-submitted: auto-generated
4075 in the messages that it generates (bounce messages and others, such as
4076 warnings). In the case of bounce messages for non-SMTP messages, there was
4077 also a typo: it was using "Auto_submitted" (underscore instead of
4078 hyphen). Since every message generated by Exim is necessarily in response
4079 to another message, thes have all been changed to:
4081 Auto-Submitted: auto-replied
4083 in accordance with these statements in the RFC:
4085 The auto-replied keyword:
4087 - SHOULD be used on messages sent in direct response to another
4088 message by an automatic process,
4090 - MUST NOT be used on manually-generated messages,
4092 - MAY be used on Delivery Status Notifications (DSNs) and Message
4093 Disposition Notifications (MDNs),
4095 - MUST NOT be used on messages generated by automatic or periodic
4096 processes, except for messages which are automatic responses to
4099 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
4100 to the default Received: header definition.
4102 PH/12 Added log selector acl_warn_skipped (default on).
4104 PH/13 After a successful wildlsearch lookup, discard the values of numeric
4105 variables because (a) they are in the wrong storage pool and (b) even if
4106 they were copied, it wouldn't work properly because of the caching.
4108 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
4109 checking when decoding. Apparently there are clients that generate
4110 overlong encoded strings. Why am I not surprised?
4112 PH/15 If the first argument of "${if match_address" was not empty, but did not
4113 contain an "@" character, Exim crashed. Now it writes a panic log message
4114 and treats the condition as false.
4116 PH/16 In autoreply, treat an empty string for "once" the same as unset.
4118 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
4119 extension "envelope-auth". The code is finished and in agreement with
4120 other implementations, but there is no documentation so far and in fact,
4121 nobody wrote the draft yet. This extension is currently #undef'ed, thus
4122 not changing the active code.
4124 Print executed "if" and "elsif" statements when debugging is used. This
4125 helps a great deal to understand what a filter does.
4127 Document more things not specified clearly in RFC3028. I had all this
4128 sorted out, when out of a sudden new issues came to my mind. Oops."
4130 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
4133 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
4134 canonical form (as documented). However, after a host name lookup from
4135 the IP address, check_host() was doing a simple string comparison with
4136 addresses acquired from the DNS when checking that the found name did
4137 have the original IP as one of its addresses. Since any found IPv6
4138 addresses are likely to be in abbreviated form, the comparison could
4139 fail. Luckily, there already exists a function for doing the comparison
4140 by converting both addresses to binary, so now that is used instead of
4141 the text comparison.
4143 PH/20 There was another similar case to PH/19, when a complete host name was
4144 given in a host list; looking up its IP address could give an abbreviated
4145 form, whereas the current host's name might or might not be abbreviated.
4146 The same fix has been applied.
4152 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
4153 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
4156 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
4157 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
4159 PH/03 Typo: missing ".o" in src/pcre/Makefile.
4161 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
4162 header line, restrict the check to what is listed in RFCs 2369 and 2929.
4163 Also, for "Auto-Submitted", treat anything other than "no" as
4164 non-personal, in accordance with RFC 3834. (Previously it treated
4165 anything starting "auto-" as non-personal.)
4167 TF/01 The control=submission/name=... option had a problem with syntax
4168 errors if the name included a slash character. The /name= option
4169 now slurps the rest of the string, so it can include any characters
4170 but it must come last in the list of options (after /sender_retain
4173 PH/05 Some modifications to the interface to the fake nameserver for the new
4181 TK/01 Added the "success_on_redirect" address verification option. See
4182 NewStuff for rationale and an example.
4184 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
4186 PH/02 Patch to exigrep to allow it to work on syslog lines.
4188 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
4189 fread() to skip over the body file's header line, because in Cygwin the
4190 header line is locked and is inaccessible.
4192 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
4193 co-exist for some time) to make it clear that it is the Exim ID that is
4194 referenced, not the Message-ID: header line.
4196 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
4197 string_format() function, because snprintf() does not exist on all
4200 PH/06 The use of forbid_filter_existstest now also locks out the use of the
4201 ${stat: expansion item.
4203 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
4204 protocol synchronization error", to keep the pedants happy.
4206 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
4207 well as for IRIX systems, when gcc is being used. See the host.c source
4210 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
4212 PH/10 Named domain lists were not working if used in a queue_smtp_domains
4215 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
4216 transport and to the smtp transport in LMTP mode.
4218 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
4220 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
4221 run a filter in a subprocess. This could lead to confusion in subsequent
4222 lookups in the parent process. There should also be a search_tidyup() at
4223 the end of the subprocess.
4225 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
4226 only if the host matched helo_try_verify_hosts, which caused the
4227 verification to occur when the EHLO/HELO command was issued. The ACL just
4228 tested the remembered result. Now, if a previous verification attempt has
4229 not happened, "verify = helo" does it there and then.
4231 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
4233 TK/03 Fix log output including CR from clamd.
4235 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
4236 exist provoked a memory error which could cause a segfault.
4238 PH/15 Installed PCRE 6.2
4240 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
4242 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
4243 of the problem. Specifically, suggested +O2 rather than +O1 for the
4246 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
4248 PH/20 If a delivery was routed to a non-standard port by means of an SRV
4249 record, the port was not correctly logged when the outgoing_port log
4250 selector was set (it logged the transort's default port).
4252 PH/21 Added support for host-specific ports to manualroute, queryprogram,
4253 fallback_hosts, and "hosts" in the smtp transport.
4255 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
4256 host errors such as "Connection refused".
4258 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
4259 authentication with radiusclient 0.4.9:
4261 - Error returned from rc_read_config was caught wrongly
4262 - Username/password not passed on to radius server due to wrong length.
4264 The presumption is that some radiusclient API changes for 4.51/PH/17
4265 were not taken care of correctly. The code is still untested by me (my
4266 Linux distribution still has 0.3.2 of radiusclient), but it was
4267 contributed by a Radius user.
4269 PH/24 When doing a callout, the value of $domain wasn't set correctly when
4270 expanding the "port" option of the smtp transport.
4272 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
4273 while reading a MIME header. Thanks to Tom Hughes for a patch.
4275 PH/24 Include config.h inside local_scan.h so that configuration settings are
4278 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
4279 that in an ACL for RCPT (for example), you can examine exactly what was
4282 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
4283 commands, but it was not correctly comparing the address with the actual
4284 client host address. Thus, it would show the EHLO address in Received:
4285 header lines when this was not necessary.
4287 PH/27 Added the % operator to ${eval:}.
4289 PH/28 Exim tries to create and chdir to its spool directory when it starts;
4290 it should be ignoring failures (because with -C, for example, it has lost
4291 privilege). It wasn't ignoring creation failures other than "already
4294 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
4297 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
4298 because an input line was too long, either on its own, or by virtue of
4299 too many continuations, the temporary file was not being removed, and the
4300 return code was incorrect.
4302 PH/31 Missing "BOOL" in function definition in filtertest.c.
4304 PH/32 Applied Sieve patches from the maintainer.
4306 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
4308 PH/33 Added "verify = not_blind".
4310 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
4311 Local/Makefile (with some defaults set). These are used in built scripts
4312 such as exicyclog, but they have never been used in the exim_install
4313 script (though there are many overriding facilities there). I have
4314 arranged that the exim_install script now takes note of these two
4317 PH/35 Installed configuration files for Dragonfly.
4319 PH/36 When a locally submitted message by a trusted user did not contain a
4320 From: header, and the sender address was obtained from -f or from an SMTP
4321 MAIL command, and the trusted user did not use -F to supply a sender
4322 name, $originator_name was incorrectly used when constructing a From:
4323 header. Furthermore, $originator_name was used for submission mode
4324 messages from external hosts without From: headers in a similar way,
4325 which is clearly wrong.
4327 PH/37 Added control=suppress_local_fixups.
4329 PH/38 When log_selector = +received_sender was set, and the addition of the
4330 sender made the log line's construction buffer exactly full, or one byte
4331 less than full, an overflow happened when the terminating "\n" was
4334 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
4335 when the result of a list match is failure because a DNS lookup failed.
4337 PH/40 RM_COMMAND is now used in the building process.
4339 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
4340 the "build-* directories that it finds.
4342 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
4343 address in a domain literal was a prefix of an interface address.
4345 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
4346 when verifying a sender address, unless rewrite_headers is false.
4348 PH/44 Wrote a long comment about why errors_to addresses are verified as
4349 recipients, not senders.
4351 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
4352 the ratelimit ACL was added.
4354 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
4356 PH/46 Added extra information about PostgreSQL errors to the error string.
4358 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
4359 avoiding the need to install special zones in a real server. This is
4360 backwards compatible; if it can't find the fake resolver, it drops back.
4361 Thus, both old and new test suites can be run.
4363 TF/02 Added util/ratelimit.pl
4365 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
4366 clock is set back in time.
4368 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
4369 Candler <B.Candler@pobox.com>.
4371 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
4372 OK for addresses that are the result of redirections.
4374 PH/48 A number of further additions for the benefit of the new test suite,
4375 including a fake gethostbyname() that interfaces to the fake DNS resolver
4378 TF/06 The fix for widen_domains has also been applied to qualify_single and
4379 search_parents which are the other dnslookup options that can cause
4382 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
4385 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
4387 TF/07 Exim produced the error message "an SRV record indicated no SMTP
4388 service" if it encountered an MX record with an empty target hostname.
4389 The message is now "an MX or SRV record indicated no SMTP service".
4391 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
4392 if the DNS of the sending site is misconfigured. This is quite a
4393 common situation. This change restores the behaviour of treating a
4394 helo verification defer as a failure.
4396 PH/51 If self=fail was set on a router, the bounce message did not include the
4397 actual error message.
4403 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
4405 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
4406 command itself ended in error, the underlying message about the transport
4407 filter timeout was being overwritten with the pipe command error. Now the
4408 underlying error message should be appended to the second error message.
4410 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
4412 PH/02 Reduce the amount of output that "make" produces by default. Full output
4413 can still be requested.
4415 PH/03 The warning log line about a condition test deferring for a "warn" verb
4416 was being output only once per connection, rather than after each
4417 occurrence (because it was using the same function as for successful
4418 "warn" verbs). This seems wrong, so I have changed it.
4420 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
4421 it should not have, which might have caused a crash in the right
4422 circumstances, but probably never did.
4424 PH/04 Installed a modified version of Tony Finch's patch to make submission
4425 mode fix the return path as well as the Sender: header line, and to
4426 add a /name= option so that you can make the user's friendly name appear
4429 TF/03 Added the control = fakedefer ACL modifier.
4431 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
4432 Mark Lowes for thorough testing.
4434 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
4436 TK/03 Merged latest SRS patch from Miles Wilton.
4438 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
4439 with the definition in sysexits.h (which is #included earlier).
4440 Fortunately, Exim does not actually use EX_OK. The code used to try to
4441 preserve the sysexits.h value, by assuming that macro definitions were
4442 scanned for macro replacements. I have been disabused of this notion,
4443 so now the code just undefines EX_OK before #including unistd.h.
4445 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
4446 in the smtp transport. When a block could not be written in a single
4447 write() function, the timeout was being re-applied to each part-write.
4448 This seems wrong - if the receiver was accepting one byte at a time it
4449 would take for ever. The timeout is now adjusted when this happens. It
4450 doesn't have to be particularly precise.
4452 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
4453 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
4455 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
4456 without a domain if the check to <postmaster@domain> fails.
4458 SC/01 Eximstats: added -xls and the ability to specify output files
4459 (patch written by Frank Heydlauf).
4461 SC/02 Eximstats: use FileHandles for outputting results.
4463 SC/03 Eximstats: allow any combination of xls, txt, and html output.
4465 SC/04 Eximstats: fixed display of large numbers with -nvr option
4467 SC/05 Eximstats: fixed merging of reports with empty tables.
4469 SC/06 Eximstats: added the -include_original_destination flag
4471 SC/07 Eximstats: removed tabs and trailing whitespace.
4473 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
4475 TK/06 MBOX spool code: Add real "From " MBOX separator line
4476 so the .eml file is really in mbox format (even though
4477 most programs do not really care). Patch from Alex Miller.
4479 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4480 The latter is generated from $received_to and is only set if the
4481 message has one envelope recipient. SA can use these headers,
4482 obviously out-of-the-box. Patch from Alex Miller.
4484 PH/08 The ${def test on a variable was returning false if the variable's
4485 value was "0", contrary to what the specification has always said!
4486 The result should be true unless the variable is empty.
4488 PH/09 The syntax error of a character other than { following "${if
4489 def:variable_name" (after optional whitespace) was not being diagnosed.
4490 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4491 accidental colon was present, for example, could give incorrect results.
4493 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4494 result is used (not including appendfile, where other changes are about
4497 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4498 This involved changing a lot of size variables from int to off_t. It
4499 should work with maildirs and everything.
4501 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4502 spamd dying while we are connected to it.
4504 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4507 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4508 function for BDB 4.3.
4510 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4512 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4513 the sources for PCRE have been re-arranged and more files are now
4516 PH/15 The code I had for printing potentially long long variables in PH/11
4517 above was not the best (it lost precision). The length of off_t variables
4518 is now inspected at build time, and an appropriate printing format (%ld
4519 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4520 to be "long long int" or "long int". This is needed for the internal
4521 formatting function string_vformat().
4523 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4524 the configuration file to be ":syslog", then the script "guesses" where
4525 the logs files are, rather than using the compiled in default. In our
4526 case the guess is not the same as the compiled default, so the script
4527 suddenly stopped working when I started to use syslog. The patch checks
4528 to see if log_file_path is "". If so, it attempts to read it from exim
4529 with no configuration file to get the compiled in version, before it
4530 falls back to the previous guessing code."
4532 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4533 implementing BATV in an Exim configuration. See NewStuff for the gory
4536 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4537 Makefile that are specific to HP-UX.
4539 PH/18 If the "use_postmaster" option was set for a recipient callout together
4540 with the "random" option, the postmaster address was used as the MAIL
4541 FROM address for the random test, but not for the subsequent recipient
4542 test. It is now used for both.
4544 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4545 patch removes a few documentation additions to RFC 3028, because the
4546 latest draft now contains them. It adds the new en;ascii-case comparator
4547 and a new error check for 8bit text in MIME parts. Comparator and
4548 require names are now matched exactly. I enabled the subaddress
4549 extension, but it is not well tested yet (read: it works for me)."
4551 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4552 rework some of the code of TK/09 above to avoid the hardwired use of
4553 "%lld" and "long long". Replaced the call to snprintf() with a call to
4556 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4557 records point to non-existent hosts", "retry timeout exceeded", and
4558 "retry time not reached for any host after a long failure period".
4560 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4561 experimental DomainKeys support:
4563 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4564 (2) On an error such as an illegally used "control", the wrong name for
4565 the control was given.
4567 These problems did NOT occur unless DomainKeys support was compiled.
4569 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4571 PH/24 Added ${if match_ip condition.
4573 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4574 databases so that it will be absolutely obvious if a crash occurs in the
4575 DB library. This is a regular occurrence (often caused by mis-matched
4578 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4579 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4580 picked up on a user's system that detects such things. There doesn't seem
4581 to be a gcc warning option for this - only an attribute that has to be
4582 put on the function's prototype. It seems that in Fedora Core 4 they have
4583 set this on a number of new functions. No doubt there will be more in due
4586 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4587 specify a transport. However, if an address that was verified by such a
4588 router was the subject of a callout, Exim crashed because it tried to
4589 read the rcpt_include_affixes from the non-existent transport. Now it
4590 just assumes that the setting of that option is false. This bug was
4591 introduced by 4.51/PH/31.
4593 PH/28 Changed -d+all to exclude +memory, because that information is very
4594 rarely of interest, but it makes the output a lot bigger. People tend to
4595 do -d+all out of habit.
4597 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4598 code in os-type was giving problems when libc.so lives in lib64, like on
4601 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4602 aren't the modern standard, and it seems that some systems' include files
4603 don't always have them. Exim was already checking for some of the newer
4604 ones like T_AAAA, and defining it itself. I've added checks for all the
4605 record types that Exim uses.
4607 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4608 not automatically generating a new one, as it is supposed to. This
4609 prevented TLS from working. If the file did exist, but contained invalid
4610 data, a new version was generated, as expected. It was only the case of a
4611 non-existent file that was broken.
4613 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4614 with a change in libdomainkeys > 0.64.
4616 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4617 from DNS. If the selector record carries the flag, it now has
4618 precedence over the domain-wide flag.
4620 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4622 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4623 the use of a port name that isn't defined in /etc/services) occurred, the
4624 message was deferred as in a normal delivery, and thus remained on the
4625 spool, instead of being failed because of the mua_wrapper setting. This
4626 is now fixed, and I tidied up some of the mua_wrapper messages at the
4629 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4630 the messages in a hash of arrays rather than using individual hashes.
4631 This is a bit cleaner and results in dramatic memory savings, albeit
4632 at a slight CPU cost.
4634 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4635 as requested by Marc Sherman.
4637 SC/10 Eximstats: added histograms for user specified patterns as requested
4640 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4642 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4643 fopen() in the content-scanning modules that did not already have it.
4649 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4650 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4652 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4654 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4656 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4657 to test Sieve filters that use "vacation".
4659 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4660 that changes the way the GnuTLS parameters are stored in the cache file.
4661 The new format can be generated externally. For backward compatibility,
4662 if the data in the cache doesn't make sense, Exim assumes it has read an
4663 old-format file, and it generates new data and writes a new file. This
4664 means that you can't go back to an older release without removing the
4667 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4668 work if there are any delivery delays because "one_time" forces the
4669 parent to be marked "delivered", so its unseen clone is never tried
4670 again. For this reason, Exim now forbids the simultaneous setting of
4673 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4674 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4675 entry if you want to know the details. The fix, however, neglected to
4676 consider the case when local delivery batching is involved. The test for
4677 "previously delivered" was not happening when checking to see if an
4678 address could be batched with a previous (undelivered) one; under
4679 certain circumstances this could lead to multiple deliveries to the same
4682 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4683 in its include files, and this causes problems building Exim.
4685 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4686 header_syntax) but Exim was just ignoring anything given after a slash.
4687 In particular, this caused confusion with an attempt to use "verify =
4688 reverse_host_lookup/defer_ok". An error is now given when options are
4689 supplied for verify items that do not have them. (Maybe reverse_host_
4690 lookup should have a defer_ok option, but that's a different point.)
4692 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4693 defined by RFC 821) to 2048, because there were problems with some AUTH
4694 commands, and RFC 1869 says the size should be increased for extended
4695 SMTP commands that take arguments.
4697 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4700 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4701 "unknown" error; now it says that the functionality isn't in the binary.
4703 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4704 an address' error message when a string expansion fails (syntax or
4705 whatever). Otherwise the password may appear in the log. Following change
4706 PH/42 below, there is no longer a chance of it appearing in a bounce
4709 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4711 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4712 of its addresses was ever tried. (Bugzilla bug #2.)
4714 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4715 the result incorrectly in the debug output. (It correctly added a newline
4716 to what was transported.)
4718 TF/01 Added $received_time.
4720 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4721 commented out examples of how to interface to a virus scanner and to
4722 SpamAssassin. Also added commented examples of av_scanner and
4723 spamd_address settings.
4725 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4726 and controls are allowed in which ACLs. There were a couple of minor
4727 errors. Some of the entries in the conditions table (which is a table of
4728 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4729 negation of where the condition IS allowed.
4731 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4733 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4734 header file does not have a version number, so I've had to invent a new
4735 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4736 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4737 radiusclient), but it was contributed by a Radius user.
4739 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4740 files or directories, for OpenSSL.
4742 PH/19 When an Exim process that is running as root has to create an Exim log
4743 file, it does so in a subprocess that runs as exim:exim so as to get the
4744 ownership right at creation (otherwise, other Exim processes might see
4745 the file with the wrong ownership). There was no test for failure of this
4746 fork() call, which would lead to the process getting stuck as it waited
4747 for a non-existent subprocess. Forks do occasionally fail when resources
4748 run out. I reviewed all the other calls to fork(); they all seem to check
4751 PH/20 When checking for unexpected SMTP input at connect time (before writing
4752 the banner), Exim was not dealing correctly with a non-positive return
4753 from the read() function. If the client had disconnected by this time,
4754 the result was a log entry for a synchronization error with an empty
4755 string after "input=" when read() returned zero. If read() returned -1
4756 (an event I could not check), uninitialized data bytes were printed.
4757 There were reports of junk text (parts of files, etc) appearing after
4760 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4762 PH/22 Added support for macro redefinition, and (re)definition in between
4763 driver and ACL definitions.
4765 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4766 forgetting to use the resulting value; it was using the unexpanded value.
4768 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4769 hadn't been configured. The fix is from Juergen Kreileder, who
4770 understands it better than I do:
4772 "Here's what I see happening with three configured cyrus_sasl
4773 authenticators configured (plain, login, cram-md5):
4775 On startup auth_cyrus_sasl_init() gets called for each of these.
4776 This means three calls to sasl_listmech() without a specified mech_list.
4777 => SASL tests which mechs of all available mechs actually work
4778 => three warnings about OTP not working
4779 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4781 With the patch, sasl_listmech() also gets called three times. But now
4782 SASL's mech_list option is set to the server_mech specified in the the
4783 authenticator. Or in other words, the answer from sasl_listmech()
4784 gets limited to just the mech you're testing for (which is different
4786 => the return list contains just 'plain' or 'login', 'cram-md5' or
4787 nothing depending on the value of ob->server_mech.
4789 I've just tested the patch: Authentication still works fine,
4790 unavailable mechs specified in the exim configuration are still
4791 caught, and the auth.log warnings about OTP are gone."
4793 PH/25 When debugging is enabled, the contents of the command line are added
4794 to the debugging output, even when log_selector=+arguments is not
4797 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4798 answer is "GNU", and only if the return is "GNU/something" is the answer
4801 PH/27 $acl_verify_message is now set immediately after the failure of a
4802 verification in an ACL, and so is available in subsequent modifiers. In
4803 particular, the message can be preserved by coding like this:
4805 warn !verify = sender
4806 set acl_m0 = $acl_verify_message
4808 Previously, $acl_verify_message was set only while expanding "message"
4809 and "log_message" when a very denied access.
4811 PH/28 Modified OS/os.c-Linux with
4813 -#ifndef OS_LOAD_AVERAGE
4814 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4816 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4817 nomenclature these days.)
4819 PH/29 Installed patch from the Sieve maintainer that adds the options
4820 sieve_useraddress and sieve_subaddress to the redirect router.
4822 PH/30 In these circumstances:
4823 . Two addresses routed to the same list of hosts;
4824 . First host does not offer TLS;
4825 . First host accepts first address;
4826 . First host gives temporary error to second address;
4827 . Second host offers TLS and a TLS session is established;
4828 . Second host accepts second address.
4829 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4830 and peerdn, if requested) that were in fact used only for the second
4833 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4834 attention to any local part prefix or suffix that was matched by the
4835 router that accepted the address. It now behaves in the same way as it
4836 does for delivery: the affixes are removed from the local part unless
4837 rcpt_include_affixes is set on the transport.
4839 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4840 timeout during the DATA phase of an incoming message.
4842 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4843 applied a patch sanctioned by the Sieve maintainer.
4845 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4846 the uid or gid is negative. A case of a negative gid caused this to be
4847 noticed. The fix allows for either to be negative.
4849 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4850 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4853 PH/36 The change PH/12 above was broken. Fixed it.
4855 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4856 the grounds that routing the same address twice would always produce the
4857 same answer. This might have been true once, but it is certainly no
4858 longer true now. Routing a child address may depend on the previous
4859 routing that produced that child. Some complicated redirection strategies
4860 went wrong when messages had multiple recipients, and made Exim's
4861 behaviour dependent on the order in which the addresses were given.
4863 I have moved the duplicate checking until after the routing is complete.
4864 Exim scans the addresses that are assigned to local and remote
4865 transports, and removes any duplicates. This means that more work will be
4866 done, as duplicates will always all be routed, but duplicates are
4867 presumably rare, so I don't expect this is of any significance.
4869 For deliveries to pipes, files, and autoreplies, the duplicate checking
4870 still happens during the routing process, since they are not going to be
4873 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4874 It corrects a timeout issue with spamd. This is Ian's comment: "The
4875 background is that sometimes spamd either never reads data from a
4876 connection it has accepted, or it never writes response data. The exiscan
4877 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4878 blindly assumes that writes won't block so it may never time out."
4880 PH/39 Allow G after quota size as well as K and M.
4882 PH/40 The value set for $authenticated_id in an authenticator may not contain
4883 binary zeroes or newlines because the value is written to log lines and
4884 to spool files. There was no check on this. Now the value is run through
4885 the string_printing() function so that such characters are converted to
4886 printable escape sequences.
4888 PH/41 $message_linecount is a new variable that contains the total number of
4889 lines in the message. Compare $body_linecount, which is the count for the
4892 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4893 bounce and delay warning messages, except in certain special cases, which
4896 (a) An SMTP error message from a remote host;
4897 (b) A message specified in a :fail: redirection;
4898 (c) A message specified in a "fail" command in a system filter;
4899 (d) A message specified in a FAIL return from the queryprogram router;
4900 (e) A message specified by the cannot_route_message router option.
4902 In these cases only, Exim does include the error details in bounce and
4903 warning messages. There are also a few cases where bland messages such
4904 as "unrouteable address" or "local delivery error" are given.
4906 PH/43 $value is now also set for the "else" part of a ${run expansion.
4908 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4909 being worked on, but at least Exim now implements the latest version to
4912 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4913 process to complete was treated as a delivery failure, a timeout while
4914 writing the message to the pipe was logged, but erroneously treated as a
4915 successful delivery. Such timeouts include transport filter timeouts. For
4916 consistency with the overall process timeout, these timeouts are now
4917 treated as errors, giving rise to delivery failures by default. However,
4918 there is now a new Boolean option for the pipe transport called
4919 timeout_defer, which, if set TRUE, converts the failures into defers for
4920 both kinds of timeout. A transport filter timeout is now identified in
4923 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4924 systems where "make" and "gmake" are different, calling "gmake" at top
4925 level broke things. I've arranged for the value of $(MAKE) to be passed
4926 from the Makefile to this script so that it can call the same version of
4930 A note about Exim versions 4.44 and 4.50
4931 ----------------------------------------
4933 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4934 changes of various kinds. As a consequence, a big documentation update was
4935 needed. This delayed the release for rather longer than seemed good, especially
4936 in the light of a couple of (minor) security issues. Therefore, the changes
4937 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4938 release. So 4.44 and 4.50 are in effect two different branches that both start
4941 I have left the 4.50 change log unchanged; it contains all the changes since
4942 4.43. The change log for 4.44 is below; many of its items are identical to
4943 those for 4.50. This seems to be the most sensible way to preserve the
4944 historical information.
4950 1. Minor wording change to the doc/README.SIEVE file.
4952 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4953 computation of the current number of files was incorrect.
4955 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4956 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4959 4. Give more explanation in the error message when the command for a transport
4960 filter fails to execute.
4962 5. There are several places where Exim runs a non-Exim command in a
4963 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4964 was being done only for the command run by the queryprogram router. It is
4965 now done for all such subprocesses. The other cases are: ${run, transport
4966 filters, and the commands run by the lmtp and pipe transports.
4968 6. Added CONFIGURE_GROUP build-time option.
4970 7. Some older OS have a limit of 256 on the maximum number of file
4971 descriptors. Exim was using setrlimit() to set 1000 as a large value
4972 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4973 systems. I've change it so that if it can't get 1000, it tries for 256.
4975 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4976 was an oversight, and furthermore, ever since the addition of extra
4977 controls (e.g. 4.43/32), the checks on when to allow different forms of
4978 "control" were broken. There should now be diagnostics for all cases when a
4979 control that does not make sense is encountered.
4981 9. Added the /retain_sender option to "control=submission".
4983 10. $recipients is now available in the predata ACL (oversight).
4985 11. Tidy the search cache before the fork to do a delivery from a message
4986 received from the command line. Otherwise the child will trigger a lookup
4987 failure and thereby defer the delivery if it tries to use (for example) a
4988 cached ldap connection that the parent has called unbind on.
4990 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4991 of $address_data from the recipient verification was clobbered by the
4992 sender verification.
4994 13. The value of address_data from a sender verification is now available in
4995 $sender_address_data in subsequent conditions in the ACL statement.
4997 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4999 15. Added a new option "connect=<time>" to callout options, to set a different
5002 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
5003 was its contents. (It was OK if the option was not defined at all.)
5005 17. A "Completed" log line is now written for messages that are removed from
5006 the spool by the -Mrm option.
5008 18. New variables $sender_verify_failure and $recipient_verify_failure contain
5009 information about exactly what failed.
5011 19. Added -dd to debug only the daemon process.
5013 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
5014 handles timeouts, both on the server side and network timeouts. Renamed the
5015 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
5017 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
5018 It is now set to "smtps".
5020 22. $host_address is now set to the target address during the checking of
5021 ignore_target_hosts.
5023 23. When checking ignore_target_hosts for an ipliteral router, no host name was
5024 being passed; this would have caused $sender_host_name to have been used if
5025 matching the list had actually called for a host name (not very likely,
5026 since this list is usually IP addresses). A host name is now passed as
5029 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5030 code that specifies a non-restarting handler (typically sigaction() in
5031 modern systems) in an attempt to fix a rare and obscure crash bug.
5033 25. Narrowed the window for a race in the daemon that could cause it to ignore
5034 SIGCHLD signals. This is not a major problem, because they are used only to
5035 wake it up if nothing else does.
5037 26. A malformed maildirsize file could cause Exim to calculate negative values
5038 for the mailbox size or file count. Odd effects could occur as a result.
5039 The maildirsize information is now recalculated if the size or filecount
5042 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5043 support for a long time. Removed HAVE_SYS_VFS_H.
5045 28. Installed the latest version of exipick from John Jetmore.
5047 29. In an address list, if the pattern was not a regular expression, an empty
5048 subject address (from a bounce message) matched only if the pattern was an
5049 empty string. Non-empty patterns were not even tested. This was the wrong
5050 because it is perfectly reasonable to use an empty address as part of a
5051 database query. An empty address is now tested by patterns that are
5052 lookups. However, all the other forms of pattern expect the subject to
5053 contain a local part and a domain, and therefore, for them, an empty
5054 address still always fails if the pattern is not itself empty.
5056 30. Exim went into a mad DNS loop when attempting to do a callout where the
5057 host was specified on an smtp transport, and looking it up yielded more
5058 than one IP address.
5060 31. Re-factored the code for checking spool and log partition space into a
5061 function that finds that data and another that does the check. The former
5062 is then used to implement four new variables: $spool_space, $log_space,
5063 $spool_inodes, and $log_inodes.
5065 32. The RFC2047 encoding function was originally intended for short strings
5066 such as real names; it was not keeping to the 75-character limit for
5067 encoded words that the RFC imposes. It now respects the limit, and
5068 generates multiple encoded words if necessary. To be on the safe side, I
5069 have increased the buffer size for the ${rfc2047: expansion operator from
5072 33. It is now permitted to omit both strings after an "if" condition; if the
5073 condition is true, the result is "true". As before, when the second string
5074 is omitted, a false condition yields an empty string. This makes it less
5075 cumbersome to write custom ACL and router conditions.
5077 34. Failure to deliver a bounce message always caused it to be frozen, even if
5078 there was an errors_to setting on the router. The errors_to setting is now
5081 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5082 canonical form (fully expanded) before being placed in
5083 $sender_host_address.
5085 36. The table in the code that translates DNS record types into text (T_A to
5086 "A" for instance) was missing entries for NS and CNAME. It is just possible
5087 that this could have caused confusion if both these types were looked up
5088 for the same domain, because the text type is used as part of Exim's
5089 per-process caching. But the chance of anyone hitting this buglet seems
5092 37. The dnsdb lookup has been extended in a number of ways.
5094 (1) There is a new type, "zns", which walks up the domain tree until it
5095 finds some nameserver records. It should be used with care.
5097 (2) There is a new type, "mxh", which is like "mx" except that it returns
5098 just the host names, not the priorities.
5100 (3) It is now possible to give a list of domains (or IP addresses) to be
5101 looked up. The behaviour when one of the lookups defers can be
5102 controlled by a keyword.
5104 (4) It is now possible to specify the separator character for use when
5105 multiple records are returned.
5107 38. The dnslists ACL condition has been extended: it is now possible to supply
5108 a list of IP addresses and/or domains to be looked up in a particular DNS
5111 39. Added log_selector=+queue_time_overall.
5113 40. When running the queue in the test harness, wait just a tad after forking a
5114 delivery process, to get repeatability of debugging output.
5116 41. Include certificate and key file names in error message when GnuTLS fails
5117 to set them up, because the GnuTLS error message doesn't include the name
5118 of the failing file when there is a problem reading it.
5120 42. Allow both -bf and -bF in the same test run.
5122 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
5124 44. The "Exiscan patch" is now merged into the mainline Exim source.
5126 45. Sometimes the final signoff response after QUIT could fail to get
5127 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5128 < 0 before doing a fflush(). This bug looks as though it goes back to the
5129 introduction of TLS in release 3.20, but "sometimes" must have been rare
5130 because the tests only now provoked it.
5132 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
5133 (this can affect the format of dates).
5135 47. exim_tidydb, when checking for the continued existence of a message for
5136 which it has found a message-specific retry record, was not finding
5137 messages that were in split spool directories. Consequently, it was
5138 deleting retry records that should have stayed in existence.
5140 48. Steve fixed some bugs in eximstats.
5142 49. The SPA authentication driver was not abandoning authentication and moving
5143 on to the next authenticator when an expansion was forced to fail,
5144 contradicting the general specification for all authenticators. Instead it
5145 was generating a temporary error. It now behaves as specified.
5147 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
5148 (the order specifies the preference for clients). The order is now AES256,
5149 AES128, 3DES, ARCFOUR128.
5151 51. Small patch to Sieve code - explicitly set From: when generating an
5154 52. Exim crashed if a remote delivery caused a very long error message to be
5155 recorded - for instance if somebody sent an entire SpamAssassin report back
5156 as a large number of 550 error lines. This bug was coincidentally fixed by
5157 increasing the size of one of Exim's internal buffers (big_buffer) that
5158 happened as part of the Exiscan merge. However, to be on the safe side, I
5159 have made the code more robust (and fixed the comments that describe what
5162 53. Now that there can be additional text after "Completed" in log lines (if
5163 the queue_time_overall log selector is set), a one-byte patch to exigrep
5164 was needed to allow it to recognize "Completed" as not the last thing in
5167 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
5168 patch that reportedly fixes this has been added. I am not expert enough to
5169 create a test for it. This is what the patch creator wrote:
5171 "I found a little strange behaviour of ldap code when working with
5172 Windows 2003 AD Domain, where users was placed in more than one
5173 Organization Units. When I tried to give exim partial DN, the exit code
5174 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
5175 But simultaneously result of request was absolutely normal ldap result,
5176 so I produce this patch..."
5178 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
5179 REFERENCE, so I have modified the code to exclude the patch when that macro
5182 55. Some experimental protocols are using DNS PTR records for new purposes. The
5183 keys for these records are domain names, not reversed IP addresses. The
5184 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5185 leaves it alone. Component reversal etc. now happens only for IP addresses.
5188 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
5190 57. Double the size of the debug message buffer (to 2048) so that more of very
5191 long debug lines gets shown.
5193 58. The exicyclog utility now does better if the number of log files to keep
5194 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
5196 59. Two changes related to the smtp_active_hostname option:
5198 (1) $smtp_active_hostname is now available as a variable.
5199 (2) The default for smtp_banner uses $smtp_active_hostname instead
5200 of $primary_hostname.
5202 60. The host_aton() function is supposed to be passed a string that is known
5203 to be a valid IP address. However, in the case of IPv6 addresses, it was
5204 not checking this. This is a hostage to fortune. Exim now panics and dies
5205 if the condition is not met. A case was found where this could be provoked
5206 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5207 components; fortuitously, this particular loophole had already been fixed
5208 by change 4.50/55 above.
5210 If there are any other similar loopholes, the new check in host_aton()
5211 itself should stop them being exploited. The report I received stated that
5212 data on the command line could provoke the exploit when Exim was running as
5213 exim, but did not say which command line option was involved. All I could
5214 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5215 running as the user.
5218 61. There was a buffer overflow vulnerability in the SPA authentication code
5219 (which came originally from the Samba project). I have added a test to the
5220 spa_base64_to_bits() function which I hope fixes it.
5223 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
5224 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
5226 63. The daemon start-up calls getloadavg() while still root for those OS that
5227 need the first call to be done as root, but it missed one case: when
5228 deliver_queue_load_max is set with deliver_drop_privilege. This is
5229 necessary for the benefit of the queue runner, because there is no re-exec
5230 when deliver_drop_privilege is set.
5232 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
5233 This has been fixed.
5235 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5236 was in use, was not putting the data itself into the right store pool;
5237 consequently, it could be overwritten for a subsequent message in the same
5238 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5241 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
5243 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
5244 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
5245 the function were treating the return as a boolean value, which happened to
5246 work because 0=false and not-0=true, but is not correct code.
5248 68. The host_aton() function was not handling scoped IPv6 addresses (those
5249 with, for example, "%eth0" on the end) correctly.
5251 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
5252 negated items (that is, ~something) in unsigned ints. Some compilers
5253 apparently mutter when there is no cast.
5255 70. If an address verification called from an ACL failed, and did not produce a
5256 user-specific message (i.e. there was only a "system" message), nothing was
5257 put in $acl_verify_message. In this situation, it now puts the system
5260 71. Change 4.23/11 added synchronization checking at the start of an SMTP
5261 session; change 4.31/43 added the unwanted input to the log line - except
5262 that it did not do this in the start of session case. It now does.
5264 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
5265 This is wrong and can cause the other end to generate a synchronization
5266 error if it is another Exim or anything else that does the synchronization
5267 check. A QUIT command is no longer sent after a timeout.
5269 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
5270 during host lookups.
5272 74. The defer_ok option of callout verification was not working if it was used
5273 when verifying addresses in header lines, that is, for this case:
5275 verify = header_sender/callout=defer_ok
5277 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
5278 those file descriptors could be used for SMTP connections. If anything
5279 wrote to stderr (the example that came up was "warn" in embedded Perl), it
5280 could be sent to the SMTP client, causing chaos. The daemon now opens
5281 stdin, stdout, and stderr to /dev/null when it puts itself into the
5284 76. Arrange for output from Perl's "warn" command to be written to Exim's main
5285 log by default. The user can override this with suitable Perl magic.
5287 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
5288 the log message when discard triggers, was not working for the DATA ACL or
5289 for the non-SMTP ACL.
5291 78. Error message wording change in sieve.c.
5293 79. If smtp_accept_max_per_host was set, the number of connections could be
5294 restricted to fewer than expected, because the daemon was trying to set up
5295 a new connection before checking whether the processes handling previous
5296 connections had finished. The check for completed processes is now done
5297 earlier. On busy systems, this bug wouldn't be noticed because something
5298 else would have woken the daemon, and it would have reaped the completed
5301 80. If a message was submitted locally by a user whose login name contained one
5302 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
5303 It caused a spool format error. I have fixed the spool reading code. A
5304 related problem was that the "from" clause in the Received: line became
5305 illegal because of the space(s). It is now covered by ${quote_local_part.
5307 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
5310 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
5312 83. Patch from Sieve maintainer to fix unterminated string problem in
5313 "vacation" handling.
5315 84. Some minor changes to the Linux configuration files to help with other
5316 OS variants using glibc.
5318 85. One more patch for Sieve to update vacation handling to latest spec.
5321 ----------------------------------------------------
5322 See the note above about the 4.44 and 4.50 releases.
5323 ----------------------------------------------------
5329 1. Change 4.43/35 introduced a bug that caused file counts to be
5330 incorrectly computed when quota_filecount was set in an appendfile
5333 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
5334 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
5337 3. Give more explanation in the error message when the command for a transport
5338 filter fails to execute.
5340 4. There are several places where Exim runs a non-Exim command in a
5341 subprocess. The SIGUSR1 signal should be disabled for these processes. This
5342 was being done only for the command run by the queryprogram router. It is
5343 now done for all such subprocesses. The other cases are: ${run, transport
5344 filters, and the commands run by the lmtp and pipe transports.
5346 5. Some older OS have a limit of 256 on the maximum number of file
5347 descriptors. Exim was using setrlimit() to set 1000 as a large value
5348 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
5349 systems. I've change it so that if it can't get 1000, it tries for 256.
5351 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
5352 was an oversight, and furthermore, ever since the addition of extra
5353 controls (e.g. 4.43/32), the checks on when to allow different forms of
5354 "control" were broken. There should now be diagnostics for all cases when a
5355 control that does not make sense is encountered.
5357 7. $recipients is now available in the predata ACL (oversight).
5359 8. Tidy the search cache before the fork to do a delivery from a message
5360 received from the command line. Otherwise the child will trigger a lookup
5361 failure and thereby defer the delivery if it tries to use (for example) a
5362 cached ldap connection that the parent has called unbind on.
5364 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
5365 of $address_data from the recipient verification was clobbered by the
5366 sender verification.
5368 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
5369 was its contents. (It was OK if the option was not defined at all.)
5371 11. A "Completed" log line is now written for messages that are removed from
5372 the spool by the -Mrm option.
5374 12. $host_address is now set to the target address during the checking of
5375 ignore_target_hosts.
5377 13. When checking ignore_target_hosts for an ipliteral router, no host name was
5378 being passed; this would have caused $sender_host_name to have been used if
5379 matching the list had actually called for a host name (not very likely,
5380 since this list is usually IP addresses). A host name is now passed as
5383 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5384 code that specifies a non-restarting handler (typically sigaction() in
5385 modern systems) in an attempt to fix a rare and obscure crash bug.
5387 15. Narrowed the window for a race in the daemon that could cause it to ignore
5388 SIGCHLD signals. This is not a major problem, because they are used only to
5389 wake it up if nothing else does.
5391 16. A malformed maildirsize file could cause Exim to calculate negative values
5392 for the mailbox size or file count. Odd effects could occur as a result.
5393 The maildirsize information is now recalculated if the size or filecount
5396 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5397 support for a long time. Removed HAVE_SYS_VFS_H.
5399 18. Updated exipick to current release from John Jetmore.
5401 19. Allow an empty sender to be matched against a lookup in an address list.
5402 Previously the only cases considered were a regular expression, or an
5405 20. Exim went into a mad DNS lookup loop when doing a callout where the
5406 host was specified on the transport, if the DNS lookup yielded more than
5409 21. The RFC2047 encoding function was originally intended for short strings
5410 such as real names; it was not keeping to the 75-character limit for
5411 encoded words that the RFC imposes. It now respects the limit, and
5412 generates multiple encoded words if necessary. To be on the safe side, I
5413 have increased the buffer size for the ${rfc2047: expansion operator from
5416 22. Failure to deliver a bounce message always caused it to be frozen, even if
5417 there was an errors_to setting on the router. The errors_to setting is now
5420 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5421 canonical form (fully expanded) before being placed in
5422 $sender_host_address.
5424 24. Updated eximstats to version 1.33
5426 25. Include certificate and key file names in error message when GnuTLS fails
5427 to set them up, because the GnuTLS error message doesn't include the name
5428 of the failing file when there is a problem reading it.
5430 26. Expand error message when OpenSSL has problems setting up cert/key files.
5433 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
5434 (this can affect the format of dates).
5436 28. exim_tidydb, when checking for the continued existence of a message for
5437 which it has found a message-specific retry record, was not finding
5438 messages that were in split spool directories. Consequently, it was
5439 deleting retry records that should have stayed in existence.
5441 29. eximstats updated to version 1.35
5442 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
5443 1.35 - bugfix such that pie charts by volume are generated correctly
5445 30. The SPA authentication driver was not abandoning authentication and moving
5446 on to the next authenticator when an expansion was forced to fail,
5447 contradicting the general specification for all authenticators. Instead it
5448 was generating a temporary error. It now behaves as specified.
5450 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
5451 (the order specifies the preference for clients). The order is now AES256,
5452 AES128, 3DES, ARCFOUR128.
5454 31. Small patch to Sieve code - explicitly set From: when generating an
5457 32. Exim crashed if a remote delivery caused a very long error message to be
5458 recorded - for instance if somebody sent an entire SpamAssassin report back
5459 as a large number of 550 error lines. This bug was coincidentally fixed by
5460 increasing the size of one of Exim's internal buffers (big_buffer) that
5461 happened as part of the Exiscan merge. However, to be on the safe side, I
5462 have made the code more robust (and fixed the comments that describe what
5465 33. Some experimental protocols are using DNS PTR records for new purposes. The
5466 keys for these records are domain names, not reversed IP addresses. The
5467 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5468 leaves it alone. Component reversal etc. now happens only for IP addresses.
5471 34. The host_aton() function is supposed to be passed a string that is known
5472 to be a valid IP address. However, in the case of IPv6 addresses, it was
5473 not checking this. This is a hostage to fortune. Exim now panics and dies
5474 if the condition is not met. A case was found where this could be provoked
5475 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5476 components; fortuitously, this particular loophole had already been fixed
5477 by change 4.50/55 or 4.44/33 above.
5479 If there are any other similar loopholes, the new check in host_aton()
5480 itself should stop them being exploited. The report I received stated that
5481 data on the command line could provoke the exploit when Exim was running as
5482 exim, but did not say which command line option was involved. All I could
5483 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5484 running as the user.
5487 35. There was a buffer overflow vulnerability in the SPA authentication code
5488 (which came originally from the Samba project). I have added a test to the
5489 spa_base64_to_bits() function which I hope fixes it.
5492 36. The daemon start-up calls getloadavg() while still root for those OS that
5493 need the first call to be done as root, but it missed one case: when
5494 deliver_queue_load_max is set with deliver_drop_privilege. This is
5495 necessary for the benefit of the queue runner, because there is no re-exec
5496 when deliver_drop_privilege is set.
5498 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5499 was in use, was not putting the data itself into the right store pool;
5500 consequently, it could be overwritten for a subsequent message in the same
5501 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5504 38. Sometimes the final signoff response after QUIT could fail to get
5505 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5506 < 0 before doing a fflush(). This bug looks as though it goes back to the
5507 introduction of TLS in release 3.20, but "sometimes" must have been rare
5508 because the tests only now provoked it.
5514 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5515 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5516 Now it returns an integer. A number of calls were still expecting a T/F
5517 return. Fortuitously, in all cases, the tests worked in OK situations,
5518 which is the norm. However, things would have gone wrong on any write
5519 failures on the smtp file descriptor. This function is used when sending
5520 messages over SMTP and also when doing verify callouts.
5522 2. When Exim is called to do synchronous delivery of a locally submitted
5523 message (the -odf or -odi options), it no longer closes stderr before doing
5526 3. Implemented the mua_wrapper option.
5528 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5530 5. Implemented the functions header_remove(), header_testname(),
5531 header_add_at_position(), and receive_remove_recipient(), and exported them
5534 6. If an ACL "warn" statement specified the addition of headers, Exim already
5535 inserted X-ACL-Warn: at the start if there was no header name. However, it
5536 was not making this test for the second and subsequent header lines if
5537 there were newlines in the string. This meant that an invalid header could
5538 be inserted if Exim was badly configured.
5540 7. Allow an ACL "warn" statement to add header lines at the start or after all
5541 the Received: headers, as well as at the end.
5543 8. Added the rcpt_4xx retry error code.
5545 9. Added postmaster_mailfrom=xxx to callout verification option.
5547 10. Added mailfrom=xxxx to the callout verification option, for verify=
5550 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5551 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5553 12. Inserted some casts to stop certain compilers warning when using pointer
5554 differences as field lengths or precisions in printf-type calls (mostly
5555 affecting debugging statements).
5557 13. Added optional readline() support for -be (dynamically loaded).
5559 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5560 same clock tick as a message's arrival, so that its received time was the
5561 same as the "first fail" time on the retry record, and that message
5562 remained on the queue past the ultimate address timeout, every queue runner
5563 would try a delivery (because it was past the ultimate address timeout) but
5564 after another failure, the ultimate address timeout, which should have then
5565 bounced the address, did not kick in. This was a "< instead of <=" error;
5566 in most cases the first failure would have been in the next clock tick
5567 after the received time, and all would be well.
5569 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5570 being recognized when the domain list was tested by the match_domain
5571 condition in an expansion string.
5573 16. Added the ${str2b64: operator.
5575 17. Exim was always calling setrlimit() to set a large limit for the number of
5576 processes, without checking whether the existing limit was already
5577 adequate. (It did check for the limit on file descriptors.) Furthermore,
5578 errors from getrlimit() and setrlimit() were being ignored. Now they are
5579 logged to the main and panic logs, but Exim does carry on, to try to do its
5580 job under whatever limits there are.
5582 18. Imported PCRE 5.0.
5584 19. Trivial typo in log message " temporarily refused connection" (the leading
5587 20. If the log selector return_path_on_delivery was set and an address was
5588 redirected to /dev/null, the delivery process crashed because it assumed
5589 that a return path would always be set for a "successful" delivery. In this
5590 case, the whole delivery is bypassed as an optimization, and therefore no
5593 21. Internal re-arrangement: the function for sending a challenge and reading
5594 a response while authentication was assuming a zero-terminated challenge
5595 string. It's now changed to take a pointer and a length, to allow for
5596 binary data in such strings.
5598 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5600 23. Exim was not respecting finduser_retries when seeking the login of the
5601 uid under which it was called; it was always trying 10 times. (The default
5602 setting of finduser_retries is zero.) Also, it was sleeping after the final
5603 failure, which is pointless.
5605 24. Implemented tls_on_connect_ports.
5607 25. Implemented acl_smtp_predata.
5609 26. If the domain in control=submission is set empty, Exim assumes that the
5610 authenticated id is a complete email address when it generates From: or
5611 Sender: header lines.
5613 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5614 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5615 chown and chgrp in /bin and hostname in /usr/bin.
5617 28. Exim was keeping the "process log" file open after each use, just as it
5618 does for the main log. This opens the possibility of it remaining open for
5619 long periods when the USR1 signal hits a daemon. Occasional processlog
5620 errors were reported, that could have been caused by this. Anyway, it seems
5621 much more sensible not to leave this file open at all, so that is what now
5624 29. The long-running daemon process does not normally write to the log once it
5625 has entered its main loop, and it closes the log before doing so. This is
5626 so that log files can straightforwardly be renamed and moved. However,
5627 there are a couple of unusual error situations where the daemon does write
5628 log entries, and I had neglected to close the log afterwards.
5630 30. The text of an SMTP error response that was received during a remote
5631 delivery was being truncated at 512 bytes. This is too short for some of
5632 the long messages that one sometimes sees. I've increased the limit to
5635 31. It is now possible to make retry rules that apply only when a message has a
5636 specific sender, in particular, an empty sender.
5638 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5639 it possible to be selective about when SMTP synchronization is enforced.
5641 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5643 32. Implemented hosts_connection_nolog.
5645 33. Added an ACL for QUIT.
5647 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5650 35. Added mailbox_size and mailbox_filecount to appendfile.
5652 36. Added control = no_multiline_responses to ACLs.
5654 37. There was a bug in the logic of the code that waits for the clock to tick
5655 in the case where the clock went backwards by a substantial amount such
5656 that the microsecond fraction of "now" was more than the microsecond
5657 fraction of "then" (but the whole seconds number was less).
5659 38. Added support for the libradius Radius client library this is found on
5660 FreeBSD (previously only the radiusclient library was supported).
5666 1. When certain lookups returned multiple values in the form name=value, the
5667 quoting of the values was not always being done properly. Specifically:
5668 (a) If the value started with a double quote, but contained no whitespace,
5670 (b) If the value contained whitespace other than a space character (i.e.
5671 tabs or newlines or carriage returns) it was not quoted.
5672 This fix has been applied to the mysql and pgsql lookups by writing a
5673 separate quoting function and calling it from the lookup code. The fix
5674 should probably also be applied to nisplus, ibase and oracle lookups, but
5675 since I cannot test any of those, I have not disturbed their existing code.
5677 2. A hit in the callout cache for a specific address caused a log line with no
5678 reason for rejecting RCPT. Now it says "Previous (cached) callout
5679 verification failure".
5681 3. There was an off-by-one bug in the queryprogram router. An over-long
5682 return line was truncated at 256 instead of 255 characters, thereby
5683 overflowing its buffer with the terminating zero. As well as fixing this, I
5684 have increased the buffer size to 1024 (and made a note to document this).
5686 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5687 when Exim is waiting for an SMTP response from a remote server, Exim
5688 restarts its select() call on the socket, thereby resetting its timeout.
5689 This is not a problem when such interrupts are rare. Somebody set up a cron
5690 job to run exiwhat every 2 minutes, which is less than the normal select()
5691 timeout (5 or 10 minutes). This meant that the select() timeout never
5692 kicked in because it was always reset. I have fixed this by comparing the
5693 time when an interrupt arrives with the time at the start of the first call
5694 to select(). If more time than the timeout has elapsed, the interrupt is
5695 treated as a timeout.
5697 5. Some internal re-factoring in preparation for the addition of Sieve
5698 extensions (by MH). In particular, the "personal" test is moved to a
5699 separate function, and given an option for scanning Cc: and Bcc: (which is
5700 not set for Exim filters).
5702 6. When Exim created an email address using the login of the caller as the
5703 local part (e.g. when creating a From: or Sender: header line), it was not
5704 quoting the local part when it contained special characters such as @.
5706 7. Installed new OpenBSD configuration files.
5708 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5709 try to make them clearer.
5711 9. Callout options, other than the timeout value, were being ignored when
5712 verifying sender addresses in header lines. For example, when using
5714 verify = header_sender/callout=no_cache
5716 the cache was (incorrectly) being used.
5718 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5719 only the Cygwin environment.
5721 11. When return_path_on_delivery was set as a log selector, if different remote
5722 addresses in the same message used different return paths and parallel
5723 remote delivery occurred, the wrong values would sometimes be logged.
5724 (Whenever a remote delivery process finished, the return path value from
5725 the most recently started remote delivery process was logged.)
5727 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5728 lines when AUTH and/or TLS are in use. This is the "received protocol"
5729 field. Exim used to use "asmtp" for authenticated SMTP, without any
5730 indication (in the protocol name) for TLS use. Now it follows the RFC and
5731 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5732 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5733 names appear in log lines as well as in Received: header lines.
5735 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5736 extensions, and comparison tests, and to fix some bugs.
5738 14. Changes to the "personal" filter test:
5740 (1) The test was buggy in that it was just doing the equivalent of
5741 "contains" tests on header lines. For example, if a user's address was
5742 anne@some.where, the "personal" test would incorrectly be true for
5744 To: susanne@some.where
5746 This test is now done by extracting each address from the header in turn,
5747 and checking the entire address. Other tests that are part of "personal"
5748 are now done using regular expressions (for example, to check local parts
5749 of addresses in From: header lines).
5751 (2) The list of non-personal local parts in From: addresses has been
5752 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5753 taken from the Sieve specification recommendations.
5755 (3) If the message contains any header line starting with "List-" it is
5756 treated as non-personal.
5758 (4) The test for "circular" in the Subject: header line has been removed
5759 because it now seems ill-conceived.
5761 15. Minor typos in src/EDITME comments corrected.
5763 16. Installed latest exipick from John Jetmore.
5765 17. If headers_add on a router specified a text string that was too long for
5766 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5767 of string_sprintf() is now avoided.
5769 18. $message_body_size was not set (it was always zero) when running the DATA
5770 ACL and the local_scan() function.
5772 19. For the "mail" command in an Exim filter, no default was being set for
5773 the once_repeat time, causing a random time value to be used if "once" was
5774 specified. (If the value happened to be <= 0, no repeat happened.) The
5775 default is now 0s, meaning "never repeat". The "vacation" command was OK
5776 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5777 (I found it when inspecting the code).
5779 20. There is now an overall timeout for performing a callout verification. It
5780 defaults to 4 times the callout timeout, which applies to individual SMTP
5781 commands during the callout. The overall timeout applies when there is more
5782 than one host that can be tried. The timeout is checked before trying the
5783 next host. This prevents very long delays if there are a large number of
5784 hosts and all are timing out (e.g. when the network connections are timing
5785 out). The value of the overall timeout can be changed by specifying an
5786 additional sub-option for "callout", called "maxwait". For example:
5788 verify = sender/callout=5s,maxwait=20s
5790 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5791 to the end before writing, but this should make it even safer).
5793 22. Exim was forgetting that it had advertised PIPELINING for the second and
5794 subsequent messages on an SMTP connection. It was also not resetting its
5795 memory on STARTTLS and an internal HELO.
5797 23. When Exim logs an SMTP synchronization error within a session, it now
5798 records whether PIPELINING has been advertised or not.
5800 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5801 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5802 rather than long int.
5804 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5806 26. Added the never_mail option to autoreply.
5812 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5813 crash if the getsockname() call failed; this can happen if a connection is
5814 closed very soon after it is established. The problem was simply in the
5815 order in which certain operations were done, causing Exim to try to write
5816 to the SMTP stream before it had set up the file descriptor. The bug has
5817 been fixed by making things happen in the correct order.
5823 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5824 before the connection was closed, thus losing the rejection response.
5826 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5827 some early Solaris releases, but causes trouble in current releases where
5828 socklen_t is defined.
5830 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5833 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5836 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5837 that releasing the top of it at the end releases what was used for sub-
5838 expansions (unless the block got too big). However, discard this block if
5839 the first thing is a variable or header, so that we can use its block when
5840 it is dynamic (useful for very large $message_headers, for example).
5842 6. Lookups now cache *every* query, not just the most recent. A new, separate
5843 store pool is used for this. It can be recovered when all lookup caches are
5844 flushed. Lookups now release memory at the end of their result strings.
5845 This has involved some general refactoring of the lookup sources.
5847 7. Some code has been added to the store_xxx() functions to reduce the amount
5848 of flapping under certain conditions.
5850 8. log_incoming_interface used to affect only the <= reception log lines. Now
5851 it causes the local interface and port to be added to several more SMTP log
5852 lines, for example "SMTP connection from", and rejection lines.
5854 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5856 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5858 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5859 could be overwritten at the end of the current message (or the start of a
5860 new message if it was set in a HELO ACL). The value is now preserved for
5861 the duration of the SMTP connection.
5863 12. If a transport had a headers_rewrite setting, and a matching header line
5864 contained an unqualified address, that address was qualified, even if it
5865 did not match any rewriting rules. The underlying bug was that the values
5866 of the flags that permit the existence of unqualified sender and recipient
5867 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5868 non-local messages, and by -bnq for local messages) were not being
5869 preserved with the message after it was received.
5871 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5872 "next input=" as part of the text comprising the host identity instead of
5873 the correct text. The code was using the same buffer for two different
5874 strings. However, depending on which order the printing function evaluated
5875 its arguments, the bug did not always show up. Under Linux, for example, my
5876 test suite worked just fine.
5878 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5879 doesn't work with some older versions of Perl. It has been changed to "my",
5880 which in any case is probably the better facility to use.
5882 15. A really picky compiler found some instances of statements for creating
5883 error messages that either had too many or two few arguments for the format
5886 16. The size of the buffer for calls to the DNS resolver has been increased
5887 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5888 for addresses that have a lot of PTR records. This alleviates a problem; it
5889 does not fully solve it.
5891 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5892 buffer now truncates the list and logs the incident, which is the same
5893 action as happens when Exim is looking up a host name and its aliases.
5894 Previously in this situation something unpredictable would happen;
5895 sometimes it was "internal error: store_reset failed".
5897 18. If a server dropped the connection unexpectedly when an Exim client was
5898 using GnuTLS and trying to read a response, the client delivery process
5899 crashed while trying to generate an error log message.
5901 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5902 string, for example:
5904 warn message = H1: something\nH2: something
5906 the text was added as a single header line from Exim's point of view
5907 though it ended up OK in the delivered message. However, searching for the
5908 second and subsequent header lines using $h_h2: did not work. This has been
5909 fixed. Similarly, if a system filter added multiple headers in this way,
5910 the routers could not see them.
5912 20. Expanded the error message when iplsearch is called with an invalid key to
5913 suggest using net-iplsearch in a host list.
5915 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5916 ACLs are no longer actually imposed (and a message to that effect is
5919 22. If a "gecos" field in a passwd entry contained escaped characters, in
5920 particular, if it contained a \" sequence, Exim got it wrong when building
5921 a From: or a Sender: header from that name. A second bug also caused
5922 incorrect handling when an unquoted " was present following a character
5923 that needed quoting.
5925 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5926 was not being matched caselessly.
5928 24. Arranged for all hyphens in the exim.8 source to be escaped with
5931 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5932 properly. Recipient callout cache records were still being keyed to include
5933 the sender, even when use_sender was set false. This led to far more
5934 callouts that were necessary. The sender is no longer included in the key
5935 when use_sender is false.
5937 26. Added "control = submission" modifier to ACLs.
5939 27. Added the ${base62d: operator to decode base 62 numbers.
5941 28. dnsdb lookups can now access SRV records.
5943 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5944 the configuration file.
5946 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5947 (-v) mode. This makes the output for a verbose queue run more intelligible.
5949 31. Added a use_postmaster feature to recipient callouts.
5951 32. Added the $body_zerocount variable, containing the number of binary zero
5952 bytes in the message body.
5954 33. The time of last modification of the "new" subdirectory is now used as the
5955 "mailbox time last read" when there is a quota error for a maildir
5958 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5960 35. Added +ignore_unknown as a special item in host lists.
5962 36. Code for decoding IPv6 addresses in host lists is now included, even if
5963 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5964 address was recognized as an IP address, but was then not correctly decoded
5965 into binary, causing unexpected and incorrect effects when compared with
5972 1. Very minor rewording of debugging text in manualroute to say "list of
5973 hosts" instead of "hostlist".
5975 2. If verify=header_syntax was set, and a header line with an unqualified
5976 address (no domain) and a large number of spaces between the end of the
5977 name and the colon was received, the reception process suffered a buffer
5978 overflow, and (when I tested it) crashed. This was caused by some obsolete
5979 code that should have been removed. The fix is to remove it!
5981 3. When running in the test harness, delay a bit after writing a bounce
5982 message to get a bit more predictability in the log output.
5984 4. Added a call to search_tidyup() just before forking a reception process. In
5985 theory, someone could use a lookup in the expansion of smtp_accept_max_
5986 per_host which, without the tidyup, could leave open a database connection.
5988 5. Added the variables $recipient_data and $sender_data which get set from a
5989 lookup success in an ACL "recipients" or "senders" condition, or a router
5990 "senders" option, similar to $domain_data and $local_part_data.
5992 6. Moved the writing of debug_print from before to after the "senders" test
5995 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5996 problems for message scanning, either using a data ACL, or using
5997 local_scan() because the Received: header was not generated till after they
5998 were called (in order to set the time as the time of reception completion).
5999 I have revised the way this works. The header is now generated after the
6000 body is received, but before the ACL or local_scan() are called. After they
6001 are run, the timestamp in the header is updated.
6007 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
6008 before starting a queue runner without re-exec. This happened only when
6009 deliver_drop_privilege was set or when the Exim user was set to root. The
6010 effect of the bug was that timeouts during subsequent deliveries caused
6011 crashes instead of being properly handled. The handler is now left at its
6012 default (and expected) setting.
6014 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
6015 message, again when deliver_drop_privilege is set or Exim is run as root.
6016 The bug described in (1) was not present in this case, but the tidying up
6017 of the other signals was missing. I have made the two cases consistent.
6019 3. The ignore_target_hosts setting on a manualroute router was being ignored
6020 for hosts that were looked up using the /MX notation.
6022 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
6025 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
6026 operated on the sender address. After changing the $sender_address to <>
6027 for the sender address verify, Exim was re-instated it as the original
6028 (before rewriting) address, but remembering that it had rewritten it, so it
6029 wasn't rewriting it again. This bug also had the effect of breaking the
6030 sender address verification caching when the sender address was rewritten.
6032 6. The ignore_target_hosts option was being ignored by the ipliteral router.
6033 This has been changed so that if the ip literal address matches
6034 ignore_target_hosts, the router declines.
6036 7. Added expansion conditions match_domain, match_address, and match_local_
6037 part (NOT match_host).
6039 8. The placeholder for the Received: header didn't have a length field set.
6041 9. Added code to Exim itself and to exim_lock to test for a specific race
6042 condition that could lead to file corruption when using MBX delivery. The
6043 issue is with the lockfile that is created in /tmp. If this file is removed
6044 after a process has opened it but before that process has acquired a lock,
6045 there is the potential for a second process to recreate the file and also
6046 acquire a lock. This could lead to two Exim processes writing to the file
6047 at the same time. The added code performs the same test as UW imapd; it
6048 checks after acquiring the lock that its file descriptor still refers to
6049 the same named file.
6051 10. The buffer for building added header lines was of fixed size, 8192 bytes.
6052 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
6055 11. Added the smtp_active_hostname option. If used, this will typically be made
6056 to depend on the incoming interface address. Because $interface_address is
6057 not set up until the daemon has forked a reception process, error responses
6058 that can happen earlier (such as "too many connections") no longer contain
6061 12. If an expansion in a condition on a "warn" statement fails because a lookup
6062 defers, the "warn" statement is abandoned, and the next ACL statement is
6063 processed. Previously this caused the whole ACL to be aborted.
6065 13. Added the iplsearch lookup type.
6067 14. Added ident_timeout as a log selector.
6069 15. Added tls_certificate_verified as a log selector.
6071 16. Added a global option tls_require_ciphers (compare the smtp transport
6072 option of the same name). This controls incoming TLS connections.
6074 17. I finally figured out how to make tls_require_ciphers do a similar thing
6075 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
6076 before starting the TLS session.
6078 18. Tabs are now shown as \t in -bP output.
6080 19. If the log selector return_path_on_delivery was set, Exim crashed when
6081 bouncing a message because it had too many Received: header lines.
6083 20. If two routers both had headers_remove settings, and the first one included
6084 a superfluous trailing colon, the final name in the first list and the
6085 first name in the second list were incorrectly joined into one item (with a
6086 colon in the middle).
6092 1. Added -C and -D options to the exinext utility, mainly to make it easier
6093 to include in the automated testing, but these could be helpful when
6094 multiple configurations are in use.
6096 2. The exinext utility was not formatting the output nicely when there was
6097 an alternate port involved in the retry record key, nor when there was a
6098 message id as well (for retries that were specific to a specific message
6099 and a specific host). It was also confused by IPv6 addresses, because of
6100 the additional colons they contain. I have fixed the IPv4 problem, and
6101 patched it up to do a reasonable job for IPv6.
6103 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
6104 delivery, the log line now contains "pipelined" if PIPELINING was used.
6106 4. An SMTP transport process used to panic and die if the bind() call to set
6107 an explicit outgoing interface failed. This has been changed; it is now
6108 treated in the same way as a connect() failure.
6110 5. A reference to $sender_host_name in the part of a conditional expansion
6111 that was being skipped was still causing a DNS lookup. This no longer
6114 6. The def: expansion condition was not recognizing references to header lines
6115 that used bh_ and bheader_.
6117 7. Added the _cache feature to named lists.
6119 8. The code for checking quota_filecount in the appendfile transport was
6120 allowing one more file than it should have been.
6122 9. For compatibility with Sendmail, the command line option
6130 and sets the incoming protocol and host name (for trusted callers). The
6131 host name and its colon can be omitted when only the protocol is to be set.
6132 Note the Exim already has two private options, -pd and -ps, that refer to
6133 embedded Perl. It is therefore impossible to set a protocol value of "d" or
6134 "s", but I don't think that's a major issue.
6136 10. A number of refactoring changes to the code, none of which should affect
6139 (a) The number of logging options was getting close to filling up the
6140 32-bit word that was used as a bit map. I have split them into two classes:
6141 those that are passed in the argument to log_write(), and those that are
6142 only ever tested independently outside of that function. These are now in
6143 separate 32-bit words, so there is plenty of room for expansion again.
6144 There is no change in the user interface or the logging behaviour.
6146 (b) When building, for example, log lines, the code previously used a
6147 macro that called string_cat() twice, in order to add two strings. This is
6148 not really sufficiently general. Furthermore, there was one instance where
6149 it was actually wrong because one of the argument was used twice, and in
6150 one call a function was used. (As it happened, calling the function twice
6151 did not affect the overall behaviour.) The macro has been replaced by a
6152 function that can join an arbitrary number of extra strings onto a growing
6155 (c) The code for expansion conditions now uses a table and a binary chop
6156 instead of a serial search (which was left over from when there were very
6157 few conditions). Also, it now recognizes conditions like "pam" even when
6158 the relevant support is not compiled in: a suitably worded error message is
6159 given if an attempt is made to use such a condition.
6161 11. Added ${time_interval:xxxxx}.
6163 12. A bug was causing one of the ddress fields not to be passed back correctly
6164 from remote delivery subprocesses. The field in question was not being
6165 subsequently used, so this caused to problems in practice.
6167 13. Added new log selectors queue_time and deliver_time.
6169 14. Might have fixed a bug in maildirsizefile handling that threw up
6170 "unexpected character" debug warnings, and recalculated the data
6171 unnecessarily. In any case, I expanded the warning message to give more
6174 15. Added the message "Restricted characters in address" to the statements in
6175 the default ACL that block characters like @ and % in local parts.
6177 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
6178 Three changes have been made:
6180 (a) There was a serious bug; a negative response to MAIL caused the whole
6181 recipient domain to be cached as invalid, thereby blocking all messages
6182 to all local parts at the same domain, from all senders. This bug has
6183 been fixed. The domain is no longer cached after a negative response to
6184 MAIL if the sender used is not empty.
6186 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
6189 (c) A new callout option, "use_sender" has been added for people who want
6190 the modified behaviour.
6196 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
6199 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
6200 indeed breaks things for older releases.
6202 3. Added additional logging to the case where there is a problem reading data
6203 from a filter that is running in a subprocess using a pipe, in order to
6204 try to track down a specific problem.
6206 4. Testing facility fudge: when running in the test harness and attempting
6207 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
6208 getting "No route to host". Convert this to a timeout.
6210 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
6213 6. Some OS don't have socklen_t but use size_t instead. This affects the
6214 fifth argument of getsockopt() amongst other things. This is now
6215 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
6216 can be set for individual OS. I have set it for SunOS5, OSF1, and
6217 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
6218 some earlier ones do not.
6220 7. Change 4.30/15 was not doing the test caselessly.
6222 8. The standard form for an IPv6 address literal was being rejected by address
6223 parsing in, for example, MAIL and RCPT commands. An example of this kind of
6224 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6225 this, as well as the form without the "IPv6" on the front (but only when
6226 address literals are enabled, of course).
6228 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
6230 10. Exim crashed if a message with an empty sender address specified by -f
6231 encountered a router with an errors_to setting. This could be provoked only
6232 by a command such as
6236 where an empty string was supplied; "<>" did not hit this bug.
6238 11. Installed PCRE release 4.5.
6240 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
6241 remained set. It is now erased.
6243 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
6244 times from message ids (which are base 36 rather than the normal 62).
6246 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
6247 were being counted as actual protocol errors, and logged if the log
6248 selector +smtp_protocol_error was set. One cannot be perfect in this test,
6249 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
6250 and DATA following a set of rejected RCPTs do not count as protocol errors.
6251 In other words, Exim assumes they were pipelined, though this may not
6252 actually be the case. Of course, in all cases the client gets an
6253 appropriate error code.
6255 15. If a lookup fails in an ACL condition, a message about the failure may
6256 be available; it is used if testing the ACL cannot continue, because most
6257 such messages specify what the cause of the deferral is. However, some
6258 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
6259 that caused an old message to be retained and used if a later statement
6260 caused a defer, replacing the real cause of the deferral.
6262 16. If an IP address had so many PTR records that the DNS lookup buffer
6263 was not large enough to hold them, Exim could crash while trying to process
6264 the truncated data. It now detects and logs this case.
6266 17. Further to 4.21/58, another change has been made: if (and only if) the
6267 first line of a message (the first header line) ends with CRLF, a bare LF
6268 in a subsequent header line has a space inserted after it, so as not to
6269 terminate the header.
6271 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
6272 unnecessarily, used atoi() instead of strtol(), and didn't check the
6273 termination when getting file sizes from file names by regex.
6275 19. Completely re-implemented the support for maildirsize files, in the light
6276 of a number of problems with the previous contributed implementation
6277 (4.30/29). In particular:
6279 . If the quota is zero, the maildirsize file is maintained, but no quota is
6282 . If the maildir directory does not exist, it is created before any attempt
6283 to write a maildirsize file.
6285 . The quota value in the file is just a cache; if the quota is changed in
6286 the transport, the new value overrides.
6288 . A regular expression is available for excluding directories from the
6291 20. The autoreply transport checks the characters in options that define the
6292 message's headers; it allows continued headers, but it was checking with
6293 isspace() after an embedded newline instead of explicitly looking for a
6296 21. If all the "regular" hosts to which an address was routed had passed their
6297 expiry times, and had not reached their retry times, the address was
6298 bounced, even if fallback hosts were defined. Now Exim should go on to try
6301 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
6302 equivalent code in the SMTP transport. Some hosts send humungous responses
6303 to HELO/EHLO, more than 1024 it seems.
6305 23. Refactoring: code in filter.c used (void *) for "any old type" but this
6306 gives compiler warnings in some environments. I've now done it "properly",
6309 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
6310 (because of problems with the built-in one) was declared to return uschar *
6311 instead of char *, causing compiler failure.
6313 25. Fixed a file descriptor leak when processing alias/forward files.
6315 26. Fixed a minor format string issue in dbfn.c.
6317 27. Typo in exim.c: ("dmbnz" for "dbmnz").
6319 28. If a filter file refered to $h_xxx or $message_headers, and the headers
6320 contained RFC 2047 "words", Exim's memory could, under certain conditions,
6323 29. When a sender address is verified, it is cached, to save repeating the test
6324 when there is more than one recipient in a message. However, when the
6325 verification involves a callout, it is possible for different callout
6326 options to be set for different recipients. It is too complicated to keep
6327 track of this in the cache, so now Exim always runs a verification when a
6328 callout is required, relying on the callout cache for the optimization.
6329 The overhead is duplication of the address routing, but this should not be
6332 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
6333 to be verified with callout=postmaster, and the main callout worked but the
6334 postmaster check failed, the verification correctly failed. However, if a
6335 subsequent RCPT command asked for sender verification *without* the
6336 postmaster check, incorrect caching caused this verification also to fail,
6339 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
6340 it was not caching the DNS options (qualify_single, search_parents) that
6341 were used when the lookup failed. A subsequent lookup with different
6342 options therefore always gave the same answer, though there were cases
6343 where it should not have. (Example: a "domains = !$mx_any" option on a
6344 dnslookup router: the "domains" option is always processed without any
6345 widening, but the router might have qualify_single set.) Now Exim uses the
6346 cached value only when the same options are set.
6348 32. Added John Jetmore's "exipick" utility to the distribution.
6350 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
6351 than a timeout (e.g. a certificate is required, and is not provided), an
6352 Exim server now closes the connection immediately. Previously it waited for
6353 the client to close - but if the client is SSL, it seems that they each
6354 wait for each other, leading to a delay before one of them times out.
6356 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
6357 maintained 0.8.x compatibility because I don't think many are using it, and
6358 it is clearly obsolete.
6360 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
6363 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
6364 client certificate was expired. A simple patch fixes this, though I don't
6365 understand the full logic of why the verify callback is called multiple
6368 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
6369 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
6370 which causes problems with some clients (such as the Certicom SSL Plus
6371 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
6372 disables the coutermeasure allowing Eudora to connect."
6374 38. Exim was not checking that a write() to a log file succeeded. This could
6375 lead to Bad Things if a log got too big, in particular if it hit a file
6376 size limit. Exim now panics and dies if it cannot write to a log file, just
6377 as it does if it cannot open a log file.
6379 39. Modified OS/Makefile-Linux so that it now contains
6381 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
6383 The two -D definitions ensure that Exim is compiled with large file
6384 support, which makes it possible to handle log files that are bigger than
6387 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
6388 instance) a domain was checked against a named list that involved a lookup,
6389 causing $domain_data to be set, then another domain was checked against the
6390 same list, then the first domain was re-checked, the value of $domain_data
6391 after the final check could be wrong. In particular, if the second check
6392 failed, it could be set empty. This bug probably also applied to
6395 41. The strip_trailing_dot option was not being applied to the address given
6396 with the -f command-line option.
6398 42. The code for reading a message's header from the spool was incrementing
6399 $received_count, but never initializing it. This meant that the value was
6400 incorrect (doubled) while delivering a message in the same process in which
6401 it was received. In the most common configuration of Exim, this never
6402 happens - a fresh exec is done - but it can happen when
6403 deliver_drop_privilege is set.
6405 43. When Exim logs an SMTP synchronization error - client data sent too soon -
6406 it now includes up to 150 characters of the unexpected data in the log
6409 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
6410 and building data strings. The size of both of these buffers was 10 000
6411 bytes - far larger than anybody would *ever* want, thought I. Needless to
6412 say, somebody hit the limit. I have increased the maximum line length to
6413 20 000 and the maximum data length of concatenated lines to 100 000. I have
6414 also fixed two bugs, because there was no checking on these buffers. Tsk,
6415 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
6416 buffer is too small.
6418 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
6419 lsearch lookups. Now it does.
6421 46. When parsing a route_list item in a manualroute router, a fixed-length
6422 buffer was used for the list of hosts. I made this 1024 bytes long,
6423 thinking that nobody would ever have a list of hosts that long. Wrong.
6424 Somebody had a whole pile of complicated expansion conditions, and the
6425 string was silently truncated, leading to an expansion error. It turns out
6426 that it is easier to change to an unlimited length (owing to other changes
6427 that have happened since this code was originally written) than to build
6428 structure for giving a limitation error. The length of the item that
6429 expands into the list of hosts is now unlimited.
6431 47. The lsearch lookup could not handle data where the length of text line was
6432 more than 4095 characters. Such lines were truncated, leading to shortened
6433 data being returned. It should now handle lines of any length.
6435 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
6436 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
6439 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
6440 replacing strings such as BIN_DIRECTORY in a source file: the replacement
6441 no longer happens in comment lines. A list of replacements is now placed
6442 at the head of all of the source files, except those whose only change is
6443 to replace PERL_COMMAND in the very first #! line.
6445 50. Replaced the slow insertion sort in queue.c, for sorting the list of
6446 messages on the queue, with a bottom-up merge sort, using code contributed
6447 by Michael Haardt. This should make operations like -bp somewhat faster on
6448 large queues. It won't affect queue runners, except when queue_run_in_order
6451 51. Installed eximstats 1.31 in the distribution.
6453 52. Added support for SRV lookups to the dnslookup router.
6455 53. If an ACL referred to $message_body or $message_body_end, the value was not
6456 reset for any messages that followed in the same SMTP session.
6458 54. The store-handling optimization for building very long strings was not
6459 differentiating between the different store pools. I don't think this
6460 actually made any difference in practice, but I've tidied it.
6462 55. While running the routers to verify a sender address, $sender_address
6463 was still set to the sender address. This is wrong, because when routing to
6464 send a bounce to the sender, it would be empty. Therefore, I have changed
6465 it so that, while verifying a sender address, $sender_address is set to <>.
6466 (There is no change to what happens when verifying a recipient address.)
6468 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
6469 target A or AAAA records (if not already returned) without resetting the
6470 qualify_single or search_parents options of the DNS resolver. These are
6471 inappropriate in this case because the targets of MX and SRV records must
6472 be FQDNs. A broken DNS record could cause trouble if it happened to have a
6473 target that, when qualified, matched something in the local domain. These
6474 two options are now turned off when doing these lookups.
6476 57. It seems that at least some releases of Reiserfs (which does not have the
6477 concept of a fixed number of inodes) returns zero and not -1 for the
6478 number of available inodes. This interacted badly with check_spool_inodes,
6479 which assumed that -1 was the "no such thing" setting. What I have done is
6480 to check that the total number of inodes is greater than zero before doing
6481 the test of how many are available.
6483 58. When a "warn" ACL statement has a log_message modifier, the message is
6484 remembered, and not repeated. This is to avoid a lot of repetition when a
6485 message has many recipients that cause the same warning to be written.
6486 However, Exim was preserving the list of already written lines for an
6487 entire SMTP session, which doesn't seem right. The memory is now reset if a
6488 new message is started.
6490 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6491 debugging output unless log_rewrite was also set.
6493 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6494 of (int)(handle) when we know that handle contains (void *)(-1).
6496 61. The Exim daemon panic-logs an error return when it closes the incoming
6497 connection. However "connection reset by peer" seems to be common, and
6498 isn't really an error worthy of noting specially, so that particular error
6501 62. When Exim is trying to find all the local interfaces, it used to panic and
6502 die if the ioctl to get the interface flags failed. However, it seems that
6503 on at least one OS (Solaris 9) it is possible to have an interface that is
6504 included in the list of interfaces, but for which you get a failure error
6505 for this call. This happens when the interface is not "plumbed" into a
6506 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6507 failure of the "get flags" call assumes that the interface is down.
6509 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6510 makes life easier for people who are doing arithmetic on fields extracted
6511 from dates, where you often get leading zeros that should not be
6512 interpreted as octal.
6514 64. Added qualify_domain to the redirect router, to override the global
6517 65. If a pathologically long header line contained very many addresses (the
6518 report of this problem mentioned 10 000) and each of them was rewritten,
6519 Exim could use up a very large amount of memory. (It kept on making new
6520 copies of the header line as it rewrote, and never released the old ones.)
6521 At the expense of a bit more processing, the header rewriting function has
6522 been changed so that it no longer eats memory in this way.
6524 66. The generation of the Received: header has been moved from the time that a
6525 message starts to be received, to the time that it finishes. The timestamp
6526 in the Received: header should now be very close to that of the <= log
6527 line. There are two side-effects of this change:
6529 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6530 logged header lines no longer include the local Received: line, because
6531 it has not yet been created. The same applies to a copy of the message
6532 that is returned to a non-SMTP sender when a message is rejected.
6534 (b) When a filter file is tested using -bf, no additional Received: header
6535 is added to the test message. After some thought, I decided that this
6538 This change does not affect the value of $received_for. It is still set
6539 after address rewriting, but before local_scan() is called.
6541 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6543 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6544 gave an unhelpful panic error message, and a defer error. I have managed to
6545 change this behaviour so that it now rejects any supplied certificate,
6546 which seems right, as the list of acceptable certificates is empty.
6548 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6549 gave an unhelpful defer error. I have not managed to make this reject any
6550 supplied certificates, but the error message it gives is "no certificate
6551 supplied", which is not helpful.
6553 70. exigrep's output now also includes lines that are not associated with any
6554 message, but which match the given pattern. Implemented by a patch from
6555 Martin Sluka, which also tidied up the Perl a bit.
6557 71. Recipient callout verification, like sender verification, was using <> in
6558 the MAIL FROM command. This isn't really the right thing, since the actual
6559 sender may affect whether the remote host accepts the recipient or not. I
6560 have changed it to use the actual sender in the callout; this means that
6561 the cache record is now keyed on a recipient/sender pair, not just the
6562 recipient address. There doesn't seem to be a real danger of callout loops,
6563 since a callout by the remote host to check the sender would use <>.
6564 [SEE ABOVE: changed after hitting problems.]
6566 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6567 temporary errors. However, in the case of such a code being given after
6568 the end of a data transmission (i.e. after ".") Exim was failing to write
6569 a retry record for the message. (Yes, there was some broken host that was
6570 actually sending 8xx at this point.)
6572 73. An unknown lookup type in a host list could cause Exim to panic-die when
6573 the list was checked. (An example that provoked this was putting <; in the
6574 middle of a list instead of at the start.) If this happened during a DATA
6575 ACL check, a -D file could be left lying around. This kind of configuration
6576 error no longer causes Exim to die; instead it causes a defer error. The
6577 incident is still logged to the main and panic logs.
6579 74. Buglet left over from Exim 3 conversion. The message "too many messages
6580 in one connection" was written to the rejectlog but not the mainlog, except
6581 when address rewriting (yes!) was being logged.
6583 75. Added write_rejectlog option.
6585 76. When a system filter was run not as root (that is, when system_filter_user
6586 was set), the values of the $n variables were not being returned to the
6587 main process; thus, they were not subsequently available in the $sn
6590 77. Added +return_path_on_delivery log selector.
6592 78. A connection timeout was being treated differently from recipients deferred
6593 when testing hosts_max_try with a message that was older than the host's
6594 retry timeout. (The host should not be counted, thus allowing all hosts to
6595 be tried at least once before bouncing.) This may have been the cause of an
6596 occasionally reported bug whereby a message would remain on the queue
6597 longer than the retry timeout, but would be bounced if a delivery was
6598 forced. I say "may" because I never totally pinned down the problem;
6599 setting up timeout/retry tests is difficult. See also the next item.
6601 79. The ultimate address timeout was not being applied to errors that involved
6602 a combination of host plus message (for example, a timeout on a MAIL
6603 command). When an address resolved to a number of possible hosts, and they
6604 were not all tried for each delivery (e.g. because of hosts_max_try), a
6605 message could remain on the queue longer than the retry timeout.
6607 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6610 81. Fixed an obscure SMTP outgoing bug which required at least the following
6611 conditions: (a) there was another message waiting for the same server;
6612 (b) the server returned 5xx to all RCPT commands in the first message so
6613 that the message was not completed; (c) the server dropped the connection
6614 or gave a negative response to the RSET that Exim sends to abort the
6615 transaction. The observed case was a dropped connection after DATA that had
6616 been sent in pipelining mode. That is, the server had advertised PIPELINING
6617 but was not implementing it correctly. The effect of the bug was incorrect
6618 behaviour, such as trying another host, and this could lead to a crash.
6624 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6625 and daemon.c were passed as pointers to ints; they should have been
6626 pointers to socklen_t variables (which are typically unsigned ints).
6628 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6631 3. Fixed a really odd bug that affected only the testing scheme; patching a
6632 certain fixed string in the binary changed the value of another string that
6633 happened to be identical to the end of the original first string.
6635 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6636 name", it returns that address as the IP address. On some operating
6637 systems (e.g. Solaris), it also passes back the IP address string as the
6638 "host name". However, on others (e.g. Linux), it passes back an empty
6639 string. Exim wasn't checking for this, and was changing the host name to an
6640 empty string, assuming it had been canonicalized.
6642 5. Although rare, it is permitted to have more than one PTR record for a given
6643 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6644 all the names associated with an address, because they do in Solaris.
6645 However, it seems that they do not in Linux for data that comes from the
6646 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6647 I found this out when I moved to a new Linux workstation and tried to run
6648 the Exim test suite.
6650 To get round this problem I have changed the code so that it now does its
6651 own call to the DNS to look up PTR records when searching for a host name.
6652 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6653 addresses that are only in /etc/hosts are still found.
6655 This behaviour is, however, controlled by an option called host_lookup_
6656 order, which defaults to "bydns:byaddr". If people want to use the other
6657 order, or indeed, just use one or the other means of lookup, they can
6658 specify it in this variable.
6660 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6661 some operating systems, this comes back from gethostbyaddr() as an empty
6662 string, and this is what Exim used to test for. However, it seems that in
6663 other systems, "." is yielded. Exim now tests for this case too.
6665 7. The values of check_spool_space and check_log_space are now held internally
6666 as a number of kilobytes instead of an absolute number of bytes. If a
6667 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6668 kilobyte. This means that much larger values can be stored.
6670 8. Exim monitor: an attempt to get the action menu when not actually pointing
6671 at a message produces an empty menu entitled "No message selected". This
6672 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6673 no entries in it ("Shell widget menu has zero width and/or height"). So I
6674 have added a single, blank menu entry in this case.
6676 9. Added ${quote_local_part.
6678 10. MIME decoding is now applied to the contents of Subject: header lines when
6681 11. Now that a reference to $sender_host_address automatically causes a reverse
6682 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6683 host lookup before query-style lookups in lists that might use this
6684 variable. This has therefore been abolished, and the "net-" prefix is no
6685 longer necessary for query-style lookups.
6687 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6688 have been a typo for LFLAGS, so it has been changed.
6690 13. The install script calls Exim with "-C /dev/null" in order to find the
6691 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6692 to be output. However, since Exim outputs its version number before the
6693 error, it didn't break the script. It just looked ugly. I fixed this by
6694 always allowing "-C /dev/null" if the caller is root.
6696 14. Ignore overlarge ACL variable number when reading spool file - insurance
6697 against a later release with more variables having written the file.
6699 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6700 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6701 this, as well as the form without the "IPv6" on the front.
6703 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6704 OS/Makefile-Darwin file.
6706 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6707 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6709 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6710 variable. However, some OS use ints for inodes. I've added cast to long int
6711 to get rid of the compiler warning.
6713 19. I had forgotten to lock out "/../" in configuration file names when
6714 ALT_CONFIG_PREFIX was set.
6716 20. Routers used for verification do not need to specify transports. However,
6717 if such a router generated a host list, and callout was configured, Exim
6718 crashed, because it could not find a port number from the (non-existent)
6719 transport. It now assumes port 25 in this circumstance.
6721 21. Added the -t option to exigrep.
6723 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6724 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6725 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6726 them is set, LOOKUP_LSEARCH is forced.
6728 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6730 24. Added sender and host information to the "rejected by local_scan()" log
6731 line; previously there was no indication of these.
6733 25. Added .include_if_exists.
6735 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6736 Linux. It turns out that not all file systems support this. Apparently some
6737 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6738 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6739 not supported on the file descriptor, is now ignored when Exim is trying to
6740 sync a directory. This applies only to Linux.
6742 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6744 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6745 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6748 29. Added support for maildirsize files from supplied patch (modified a bit).
6750 30. The use of :fail: followed by an empty string could lead Exim to respond to
6751 sender verification failures with (e.g.):
6753 550 Verification failed for <xxx>
6754 550 Sender verify failed
6756 where the first response line was missing the '-' that indicates it is not
6757 the final line of the response.
6759 31. The loop for finding the name of the user that called Exim had a hardwired
6760 limit of 10; it now uses the value of finduser_retries, which is used for
6761 all other user lookups.
6763 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6766 33. Exim was neglecting to zero errno before one call of strtol() when
6767 expanding a string and expecting an integer value. On some systems this
6768 resulted in spurious "integer overflow" errors. Also, it was casting the
6769 result into an int without checking.
6771 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6772 did not work. The code looks as if it has *never* worked, though it appears
6773 to have been documented since at least release 1.62. I have made it work.
6775 35. The "timeout_DNS" error in retry rules, also documented since at least
6776 1.62, also never worked. As it isn't clear exactly what this means, and
6777 clearly it isn't a major issue, I have abolished the feature by treating it
6778 as "timeout", and writing a warning to the main and panic logs.
6780 36. The display of retry rules for -brt wasn't always showing the error code
6783 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6784 timeout_connect_A, timeout_connect_MX.
6786 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6787 to the empty sender.
6789 39. The daemon was not analysing the content of -oX till after it had closed
6790 stderr and disconnected from the controlling terminal. This meant that any
6791 syntax errors were only noted on the panic log, and the return code from
6792 the command was 0. By re-arranging the code a little, I've made the
6793 decoding happen first, so such errors now appear on stderr, and the return
6794 code is 1. However, the actual setting up of the sockets still happens in
6795 the disconnected process, so errors there are still only recorded on the
6798 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6799 connections (as happens on some IP stacks) was logged at start up time as
6800 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6801 it from "IPv6 and IPv4", which means that two separate sockets are being
6804 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6805 says whether AF_INET or AF_INET6 was passed as an argument.
6807 42. Exiwhat output was messed up when time zones were included in log
6810 43. Exiwhat now gives more information about the daemon's listening ports,
6811 and whether -tls-on-connect was used.
6813 44. The "port" option of the smtp transport is now expanded.
6815 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6816 silently ignored. Now an error message is written to the main and panic
6819 46. There's a new ACL modifier called "logwrite" which writes to a log file
6820 as soon as it is encountered.
6822 47. Added $local_user_uid and $local_user_gid at routing time.
6824 48. Exim crashed when trying to verify a sender address that was being
6827 49. Exim was recognizing only a space character after ".include". It now also
6828 recognizes a tab character.
6830 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6831 extracting the relevant information from the specification. The man page no
6832 longer contains scrambled data for the -d option, and I've added a section
6833 at the front about calling Exim under different names.
6835 51. Added "extra_headers" argument to the "mail" command in filter files.
6837 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6840 53. Installed eximstats 1.29.
6842 54. Added transport_filter_timeout as a generic transport option.
6844 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6845 Cc: header lines. This was required by RFC 822, but it not required by RFC
6848 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6849 incoming messages that did not have them. Now it does so only if the
6850 message originates locally, that is, if there is no associated remote host
6851 address. When Resent- header lines are present, this applies to the Resent-
6852 lines rather than the non-Resent- lines.
6854 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6855 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6857 58. Messages for configuration errors now include the name of the main
6858 configuration file - useful now that there may be more than one file in a
6859 list (.included file names were always shown).
6861 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6862 for those rare installations that do not start the daemon as root or run it
6863 setuid root. I've cut out the call to initgroups() if the daemon is not
6866 60. The Exim user and group can now be bound into the binary as text strings
6867 that are looked up at the start of Exim's processing.
6869 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6871 62. Added $mailstore_basename variable.
6873 63. Installed patch to sieve.c from Michael Haardt.
6875 64. When Exim failed to open the panic log after failing to open the main log,
6876 the original message it was trying to log was written to stderr and debug
6877 output, but if they were not available (the usual case in production), it
6878 was lost. Now it is written to syslog before the two lines that record the
6879 failures to open the logs.
6881 65. Users' Exim filters run in subprocesses under the user's uid. It is
6882 possible for a "deliver" command or an alias in a "personal" command to
6883 provoke an address rewrite. If logging of address rewriting is configured,
6884 this fails because the process is not running as root or exim. There may be
6885 a better way of dealing with this, but for the moment (because 4.30 needs
6886 to be released), I have disabled address rewrite logging when running a
6887 filter in a non-root, non-exim process.
6893 1. The buildconfig auxiliary program wasn't quoting the value set for
6894 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6895 not defined. This bug was masked in 4.22 by the effect that was fixed in
6898 2. Some messages that were rejected after a message id was allocated were
6899 shown as "incomplete" by exigrep. It no longer does this for messages that
6900 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6902 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6903 have allow_domain_literals set, the ID did not get logged in the <= line.
6904 Domain literals are now always recognized in Message-ID: header lines.
6906 4. The first argument for a ${extract expansion item is the key name or field
6907 number. Leading and trailing spaces in this item were not being ignored,
6908 causing some misleading effects.
6910 5. When deliver_drop_privilege was set, single queue runner processes started
6911 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6912 same command in the process it spins off) were not dropping privilege.
6914 6. When the daemon running as "exim" started a queue runner, it always
6915 re-executed Exim in the spun-off process. This is a waste of effort when
6916 deliver_drop_privilege is set. The new process now just calls the
6917 queue-runner function directly.
6923 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6926 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6927 ignored. Though the use of -oP was forcing the writing of a pid file, it
6928 was always written to the default place.
6930 3. If the message "no IP address found for host xxxx" is generated during
6931 incoming verification, it is now followed by identification of the incoming
6932 connection (so you can more easily find what provoked it).
6934 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6936 5. Added some features to "harden" Exim a bit more against certain attacks:
6938 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6939 be put in Local/Makefile. This is like the never_users runtime option,
6940 but it cannot be overridden. The default setting is "root".
6942 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6943 prefix string with which any file named in a -C command line option
6946 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6947 is retained for -C and -D only if the caller of Exim is root. Without
6948 it, the exim user may also use -C and -D and retain privilege.
6950 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6951 command line option is disabled.
6953 6. Macro names set by the -D option must start with an upper case letter, just
6954 like macro names defined in the configuration file.
6956 7. Added "dereference=" facility to LDAP.
6958 8. Two instances of the typo "uknown" in the source files are fixed.
6960 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6961 the Configure-Makefile script screwed up while processing it.
6963 10. Incorporated PCRE 4.4.
6965 11. The SMTP synchronization check was not operating right at the start of an
6966 SMTP session. For example, it could not catch a HELO sent before the client
6967 waited for the greeting. There is now a check for outstanding input at the
6968 point when the greeting is written. Because of the duplex, asynchronous
6969 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6970 way, but not yet received, when the check is performed.
6972 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6973 on TCP/IP sockets, because this apparently causes some broken clients to
6976 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6977 unchanged) from the Cygwin maintainer.
6979 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6980 was supported instead of testing for mbx. Effectively a typo.
6982 15. The spa authenticator server code was not checking that the input it
6983 received was valid base64.
6985 16. The debug output line for the "set" modifier in ACLs was not showing the
6986 name of the variable that was being set.
6988 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6990 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6991 lookup on its own. Something else was needed to trigger the lookup. For
6992 example, a match in host_lookup or the need for a host name in a host list.
6993 Now, if $sender_host_name is referenced and the host name has not yet been
6994 looked up, a lookup is performed. If the lookup fails, the variable remains
6995 empty, and $host_lookup_failed is set to "1".
6997 19. Added "eqi" as a case-independent comparison operator.
6999 20. The saslauthd authentication condition could segfault if neither service
7000 nor realm was specified.
7002 21. If an overflowing value such as "2048M" was set for message_size_limit, the
7003 error message that was logged was misleading, and incoming SMTP
7004 connections were dropped. The message is now more accurate, and temporary
7005 errors are given to SMTP connections.
7007 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
7008 (except RSET) with a 421 error, until QUIT is received. However, it was
7009 failing to send a response to QUIT.
7011 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
7012 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
7013 ACL is now run after the helo_try_verify_hosts code.
7015 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
7016 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
7017 case-independent, so other case variants are also recognized). Apparently
7018 some systems use these upper case variants.
7020 25. If more than two messages were waiting for the same host, and a transport
7021 filter was specified for the transport, Exim sent two messages over the
7022 same TCP/IP connection, and then failed with "socket operation on non-
7023 socket" when it tried to send the third.
7025 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
7027 27. The extern definition of crypt16() in expand.c was not being excluded when
7028 the OS had its own crypt16() function.
7030 28. Added bounce_return_body as a new option, and bounce_return_size_limit
7031 as a preferred synonym for return_size_limit, both as an option and as an
7034 29. Added LIBS=-liconv to OS/Makefile-OSF1.
7036 30. Changed the default configuration ACL to relax the local part checking rule
7037 for addresses that are not in any local domains. For these addresses,
7038 slashes and pipe symbols are allowed within local parts, but the sequence
7039 /../ is explicitly forbidden.
7041 31. SPA server authentication was not clearing the challenge buffer before
7044 32. log_message in a "warn" ACL statement was writing to the reject log as
7045 well as to the main log, which contradicts the documentation and doesn't
7046 seem right (because no rejection is happening). So I have stopped it.
7048 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
7049 However, I am unable to do any testing of this.
7051 34. Fixed an infelicity in the appendfile transport. When checking directories
7052 for a mailbox, to see if any needed to be created, it was accidentally
7053 using path names with one or more superfluous leading slashes; tracing
7054 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
7056 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
7057 contents are added to the log line that is written for every discarded
7058 recipient. (Previously a log_message setting was ignored.)
7060 36. The ${quote: operator now quotes the string if it is empty.
7062 37. The install script runs exim in order to find its version number. If for
7063 some reason other than non-existence or emptiness, which it checks, it
7064 could not run './exim', it was installing it with an empty version number,
7065 i.e. as "exim-". This error state is now caught, and the installation is
7068 38. An argument was missing from the function that creates an error message
7069 when Exim fails to connect to the socket for saslauthd authentication.
7070 This could cause Exim to crash, or give a corrupted message.
7072 39. Added isip, isip4, and isip6 to ${if conditions.
7074 40. The ACL variables $acl_xx are now saved with the message, and can be
7075 accessed later in routers, transports, and filters.
7077 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
7078 strings in the file are not string-expanded.
7080 42. If a MAIL command specified a SIZE value that was too large to fit into an
7081 int variable, the check against message_size_limit failed. Such values are
7082 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
7083 day this will have to be increased, but I don't think I want to be around
7084 when emails are that large.
7091 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
7092 iconv() is not standard in FreeBSD.
7094 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
7095 IPv6 enabled. The observed symptom was a segmentation fault on return from
7096 the function os_common_find_running_interfaces() in src/os.c.
7098 3. In the check_special_case() function in daemon.c I had used "errno" as an
7099 argument name, which causes warnings on some systems. This was basically a
7100 typo, since it was named "eno" in the comments!
7102 4. The code that waits for the clock to tick (at a resolution of some fraction
7103 of a second) so as to ensure message-id uniqueness was always waiting for
7104 at least one whole tick, when it could have waited for less. [This is
7105 almost certainly not relevant at current processor speeds, where it is
7106 unlikely to ever wait at all. But we try to future-proof.]
7108 5. The function that sleeps for a time interval that includes fractions of a
7109 second contained a race. It did not block SIGALRM between setting the
7110 timer, and suspending (a couple of lines later). If the interval was short
7111 and the sigsuspend() was delayed until after it had expired, the suspension
7112 never ended. On busy systems this could lead to processes getting stuck for
7115 6. Some uncommon configurations may cause a lookup to happen in a queue runner
7116 process, before it forks any delivery processes. The open lookup caching
7117 mechanism meant that the open file or database connection was passed into
7118 the delivery process. The problem was that delivery processes always tidy
7119 up cached lookup data. This could cause a problem for the next delivery
7120 process started by the queue runner, because the external queue runner
7121 process does not know about the closure. So the next delivery process
7122 still has data in the lookup cache. In the case of a file lookup, there was
7123 no problem because closing a file descriptor in a subprocess doesn't affect
7124 the parent. However, if the lookup was caching a connection to a database,
7125 the connection was closed, and the second delivery process was likely to
7126 see errors such as "PGSQL: query failed: server closed the connection
7127 unexpectedly". The problem has been fixed by closing all cached lookups
7128 in a queue runner before running a delivery process.
7130 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
7131 seem to have the "const" qualifier which it has on other OS. I've
7134 8. Change 4.21/2 was too strict. It is only if there are two authenticators
7135 *of the same type* (client or server) with the same public name that an
7136 error should be diagnosed.
7138 9. When Exim looked up a host name for an IP address, but failed to find the
7139 original IP address when looking up the host name (a safety check), it
7140 output the message "<ip address> does not match any IP for NULL", which was
7141 confusing, to say the least. The bug was that the host name should have
7142 appeared instead of "NULL".
7144 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
7145 user that is built into the binary, and the -C or -D options is used, root
7146 privilege is dropped before the configuration file is read. In addition,
7147 logging is switched to stderr instead of the normal log files. If the
7148 configuration then re-defines the Exim user, the unprivileged environment
7149 is probably not what is expected, so Exim logs a panic warning message (but
7152 However, if deliver_drop_privilege is set, the unprivileged state may well
7153 be exactly what is intended, so the warning has been cut out in that case,
7154 and Exim is allowed to try to write to its normal log files.
7160 1. smtp_return_error_details was not giving details for temporary sender
7161 or receiver verification errors.
7163 2. Diagnose a configuration error if two authenticators have the same public
7166 3. Exim used not to create the message log file for a message until the first
7167 delivery attempt. This could be confusing when incoming messages were held
7168 for policy or load reasons. The message log file is now created at the time
7169 the message is received, and an initial "Received" line is written to it.
7171 4. The automatically generated man page for command line options had a minor
7172 bug that caused no ill effects; however, a more serious problem was that
7173 the procedure for building the man page automatically didn't always
7174 operate. Consequently, release 4.20 contains an out-of-date version. This
7175 shouldn't happen again.
7177 5. When building Exim with embedded Perl support, the script that builds the
7178 Makefile was calling 'perl' to find its compile-time parameters, ignoring
7179 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
7181 6. The freeze_tell option was not being used for messages that were frozen on
7182 arrival, either by an ACL or by local_scan().
7184 7. Added the smtp_incomplete_transaction log selector.
7186 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
7187 was accepting AUTH without a new EHLO.
7189 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
7190 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
7193 10. Logging of TCP/IP connections (when configured) now happens in the main
7194 daemon process instead of the child process, so that the TCP/IP connection
7195 count is more accurate (but it can never be perfect).
7197 11. The use of "drop" in a nested ACL was not being handled correctly in the
7198 outer ACL. Now, if condition failure induced by the nested "drop" causes
7199 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
7200 or "require"), the connection is dropped.
7202 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
7203 that yield "discard" can now be used with an "accept" or a "discard" verb,
7204 but an error is generated for any others (because I can't see a useful way
7205 to define what should happen).
7207 13. When an ACL is read dynamically from a file (or anywhere else), the lines
7208 are now processed in the same way as lines in the Exim configuration file.
7209 In particular, continuation lines are supported.
7211 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
7213 15. Added -ti meaning -t -i.
7215 16. Check for letters, digits, hyphens, and dots in the names of dnslist
7216 domains, and warn by logging if others are found.
7218 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
7219 returned from GIFCONF when Exim is trying to find the list of interfaces on
7220 a host. The code in os.c has been modified to copy each ifreq to an aligned
7221 structure in all cases.
7223 Also, in some cases, the returned ifreq's were being copied to a 'struct
7224 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
7225 means the last couple of bytes of an IPv6 address could be chopped if the
7226 ifreq contained only a normal sockaddr (14 bytes storage).
7228 18. Named domain lists were not supported in the hosts_treat_as_local option.
7229 An entry such as +xxxx was not recognized, and was treated as a literal
7232 19. Ensure that header lines added by a DATA ACL are included in the reject log
7233 if the ACL subsequently rejects the message.
7235 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
7236 MD5 (which is deprecated).
7238 21. When testing a filter file using -bf, Exim was writing a message when it
7239 took the sender from a "From " line in the message, but it was not doing so
7240 when it took $return_path from a Return-Path: header line. It now does.
7242 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
7243 with a valid header line field name (a series of printing characters
7244 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
7246 23. Changed "disc" in the source to "disk" to conform to the documentation and
7247 the book and for uniformity.
7249 24. Ignore Sendmail's -Ooption=value command line item.
7251 25. When execve() failed while trying to run a command in a pipe transport,
7252 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
7253 could be confused with a return value of 69 from the command itself. This
7254 has been changed to 127, the value the shell returns if it is asked to run
7255 a non-existent command. The wording for the related log line suggests a
7256 non-existent command as the problem.
7258 26. If received_header_text expands to an empty string, do not add a Received:
7259 header line to the message. (Well, it adds a token one on the spool, but
7260 marks it "old" so that it doesn't get used or transmitted.)
7262 27. Installed eximstats 1.28 (addition of -nt option).
7264 28. There was no check for failure on the call to getsockname() in the daemon
7265 code. This can fail if there is a shortage of resources on the system, with
7266 ENOMEM, for example. A temporary error is now given on failure.
7268 29. Contrary to the C standard, it seems that in some environments, the
7269 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
7270 program. Exim now does this explicitly; it affects the formatting of
7271 timestamps using strftime().
7273 30. If exiqsumm was given junk data, it threw up some uninitialized variable
7274 complaints. I've now initialized all the variables, to avoid this.
7276 32. Header lines added by a system filter were not being "seen" during
7277 transport-time rewrites.
7279 33. The info_callback() function passed to OpenSSL is set up with type void
7280 (*)(SSL *, int, int), as described somewhere. However, when calling the
7281 function (actually a macro) that sets it up, the type void(*)() is
7282 expected. I've put in a cast to prevent warnings from picky compilers.
7284 34. If a DNS black list lookup found a CNAME record, but there were no A
7285 records associated with the domain it pointed at, Exim crashed.
7287 35. If a DNS black list lookup returned more than one A record, Exim ignored
7288 all but the first. It now scans all returned addresses if a particular IP
7289 value is being sought. In this situation, the contents of the
7290 $dnslist_value variable are a list of all the addresses, separated by a
7293 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
7294 to accept a host name and all its aliases if the forward lookup for any of
7295 them yielded the IP address of the incoming connection. Now it accepts only
7296 those names whose forward lookup yields the correct IP address. Any other
7297 names are discarded. This closes a loophole whereby a rogue DNS
7298 administrator could create reverse DNS records to break through a
7299 wildcarded host restriction in an ACL.
7301 37. If a user filter or a system filter that ran in a subprocess used any of
7302 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
7303 the wrong values were passed to the pipe command ($thisaddress had the
7304 value of $0, $0 had the value of $1, etc). This bug was introduced by
7305 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
7307 38. Improved the line breaking for long SMTP error messages from ACLs.
7308 Previously, if there was no break point between 40 and 75 characters, Exim
7309 left the rest of the message alone. Two changes have been made: (a) I've
7310 reduced the minimum length to 35 characters; (b) if it can't find a break
7311 point between 35 and 75 characters, it looks ahead and uses the first one
7312 that it finds. This may give the occasional overlong line, but at least the
7313 remaining text gets split now.
7315 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
7316 file descriptors might be low, and that setting 1000 would always raise it.
7317 It turns out that in some environments, the limit is already over 1000 and
7318 that lowering it causes trouble. So now Exim takes care not to decrease it.
7320 40. When delivering a message, the value of $return_path is set to $sender_
7321 address at the start of routing (routers may change the value). By an
7322 oversight, this default was not being set up when an address was tested by
7323 -bt or -bv, which affected the outcome if any router or filter referred to
7326 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
7327 log only when "message" or "log_message" are set. However, if one of the
7328 conditions was an address verification, or a call to a nested ACL, the
7329 messages generated by the underlying test were being passed through. This
7330 no longer happens. The underlying message is available in $acl_verify_
7331 message for both "message" and "log_message" expansions, so it can be
7332 passed through if needed.
7334 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
7335 new expansion $bh_ to give the encoded byte string without charset
7336 translation. Translation happens only if iconv() is available; HAVE_ICONV
7337 indicates this at build time. HEADERS_CHARSET gives the charset to
7338 translate to; headers_charset can change it in the configuration, and
7339 "headers charset" can change it in an individual filter file.
7341 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
7342 that creates RFC 2047 encoded "words" labels them as that charset instead
7343 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
7344 expansion operator; (ii) when Exim creates a From: line for a local
7345 message; (iii) when a header line is rewritten to include a "phrase" part.
7347 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
7348 buggy, causing it to skip the first lines of messages whose message ID
7349 ended in 'D'. This would not have bitten before Exim release 4.14, because
7350 message IDs were unlikely to end in 'D' before then. The effect was to have
7351 incorrect size information for certain domains.
7353 45. #include "config.h" was missing at the start of the crypt16.c module. This
7354 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
7357 46. If there was a timeout during a "random" callout check, Exim treated it as
7358 a failure of the random address, and carried on sending RSET and the real
7359 address. If the delay was just some slowness somewhere, the response to the
7360 original RCPT would be taken as a response to RSET and so on, causing
7361 mayhem of various kinds.
7363 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
7364 when I implemented it. It didn't allow for the fact that some option values
7365 may legitimately be negative (e.g. size_addition), and it didn't even do
7366 the right test for positive values.
7368 48. Domain names in DNS records are case-independent. Exim always looks them up
7369 in lower case. Some resolvers return domain names in exactly the case they
7370 appear in the zone file, that is, they may contain uppercase letters. Not
7371 all resolvers do this - some return always lower case. Exim was treating a
7372 change of case by a resolver as a change of domain, similar to a widening
7373 of a domain abbreviation. This triggered its re-routing code and so it was
7374 trying to route what was effectively the same domain again. This normally
7375 caused routing to fail (because the router wouldn't handle the domain
7376 twice). Now Exim checks for this case specially, and just changes the
7377 casing of the domain that it ultimately uses when it transmits the message
7380 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
7383 50. If a filter generated a file delivery with a non-absolute name (possible if
7384 no home directory exists for the router), the forbid_file option was not
7387 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
7388 the existing equality matching.
7390 52. Exim was using ints instead of ino_t variables in some places where it was
7391 dealing with inode numbers.
7393 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
7394 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
7395 called TMPDIR, and if it finds it is different, it changes its value.
7397 54. The smtp_printf() function is now made available to local_scan() so
7398 additional output lines can be written before returning. There is also an
7399 smtp_fflush() function to enable the detection of a dropped connection.
7400 The variables smtp_input and smtp_batched_input are exported to
7403 55. Changed the default runtime configuration: the message "Unknown user"
7404 has been removed from the ACL, and instead placed on the localuser router,
7405 using the cannot_route_message feature. This means that any verification
7406 failures that generate their own messages won't get overridden. Similarly,
7407 the "Unrouteable address" message that was in the ACL for unverifiable
7408 relay addresses has also been removed.
7410 56. Added hosts_avoid_esmtp to the smtp transport.
7412 57. The exicyclog script was not checking for the esoteric option
7413 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
7414 will work only if exicyclog is run under the appropriate euid.
7416 58. Following a discussion on the list, the rules by which Exim recognises line
7417 endings on incoming messages have been changed. The -dropcr and drop_cr
7418 options are now no-ops, retained only for backwards compatibility. The
7419 following line terminators are recognized: LF CRLF CR. However, special
7420 processing applies to CR:
7422 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
7423 nor a local message in the state where . is a terminator.
7425 (ii) If a bare CR is encountered in a header line, an extra space is added
7426 after the line terminator so as not to end the header. The reasoning
7427 behind this is that bare CRs in header lines are most likely either
7428 to be mistakes, or people trying to play silly games.
7430 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
7431 was being incorrectly given as 18 bytes larger than it should have been.
7432 This is a VOB (very old bug).
7434 60. This may never have affected anything current, but just in case it has:
7435 When the local host is found other than at the start of a list of hosts,
7436 the local host, those with the same MX, and any that follow, are discarded.
7437 When the list in question was part of a longer list of hosts, the following
7438 hosts (not currently being processed) were also being discarded. This no
7439 longer happens. I'm not sure if this situation could ever has previously
7442 61. Added the "/MX" feature to lists of hosts in the manualroute and query
7445 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
7446 header. This is something that is recommended in a new Internet Draft, and
7447 is something that is documented as being done by Sendmail. There are two
7448 possible values. For messages generated by the autoreply transport, Exim
7451 Auto-Submitted: auto-replied
7453 whereas for all other generated messages (e.g. bounces) it adds
7455 Auto-Submitted: auto-generated
7457 63. The "personal" condition in filters now includes a test for the
7458 Auto-Submitted: header. If it contains the string "auto-" the message it
7459 not considered personal.
7461 64. Added rcpt_include_affixes as a generic transport option.
7463 65. Added queue_only_override (default true).
7465 66. Added the syslog_duplication option.
7467 67. If what should have been the first header line of a message consisted of
7468 a space followed by a colon, Exim was mis-interpreting it as a header line.
7469 It isn't of course - it is syntactically invalid and should therefore be
7470 treated as the start of the message body. The misbehaviour could have
7471 caused a number of strange effects, including loss of data in subsequent
7472 header lines, and spool format errors.
7474 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
7475 client host had authenticated. This control can now be exercised by an ACL
7476 for more flexibility.
7478 69. By default, callouts do not happen when testing with -bh. There is now a
7479 variant, -bhc, which does actually run the callout code, including
7480 consulting and updating the callout cache.
7482 70. Added support for saslauthd authentication, courtesy of Alexander
7485 71. If statvfs() failed on the spool or log directories while checking their
7486 size for availability, Exim confusingly gave the error "space shortage".
7487 Furthermore, in debugging mode it crashed with a floating point exception.
7488 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7489 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7490 serious problem, Exim now writes to the main and panic logs when this
7491 happens, with details of the failure. It then refuses to accept the
7492 incoming message, giving the message "spool directory problem" or "log
7493 directory problem" with a 421 code for SMTP messages.
7495 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7496 0, 1, and 2 exist, because some OS complain for execs without them (see
7497 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7498 descriptors. However, the code omitted to check that the open succeeded,
7499 causing mysterious errors if for some reason the permissions on /dev/null
7500 got screwed. Now Exim writes a message to the main and panic logs, and
7501 bombs out if it can't open /dev/null.
7503 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7504 interact so that it is all more flexible. It is supposed to remain
7505 backwards compatible. Also added extra_local_interfaces.
7507 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7508 to bomb out with an assertion failure - to the client this appears as a
7509 connection drop. This problem occurs in the part of the code that was taken
7510 from the Samba project. Fortunately, the assertion is in a very simple
7511 function, so I have fixed this by reproducing the function inline in the
7512 one place where it is called, and arranging for authentication to fail
7513 instead of killing the process with assert().
7515 75. The SPA client code was not working when the server requested OEM rather
7516 than Unicode encoding.
7518 76. Added code to make require_files with a specific uid setting more usable in
7519 the case where statting the file as root fails - usually a non-root-mounted
7520 NFS file system. When this happens and the failure is EACCES, Exim now
7521 forks a subprocess and does the per-uid checking as the relevant uid.
7523 77. Added process_log_path.
7525 78. If log_file_path was not explicitly set, a setting of check_log_space or
7526 check_log_inodes was ignored.
7528 79. If a space check for the spool or log partitions fails, the incident is now
7529 logged. Of course, in the latter case the data may get lost...
7531 80. Added the %p formatting code to string_format() so that it can be used to
7532 print addresses in debug_print(). Adjusted all the address printing in the
7533 debugging in store.c to use %p rather than %d.
7535 81. There was a concern that a line of code in smtp_in.c could overflow a
7536 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7537 initially expressed, the concern was not well-founded, because trailing
7538 spaces are removed early. However, if the trailing spaces were followed by
7539 a NULL, they did not get removed, so the overflow was possible. Two fixes
7542 (a) I re-wrote the offending code in a cleaner fashion.
7543 (b) If an incoming SMTP command contains a NULL character, it is rejected
7546 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7547 runs initgroups(), so that if the Exim user is in any additional groups,
7548 they will be used during message reception.
7554 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / jgh / exim.git / blob