1 This document describes *changes* to previous versions, that might
2 affect Exim's operation, with an unchanged configuration file. For new
3 options, and new features, see the NewStuff file next to this ChangeLog.
9 JH/01 OpenSSL: With debug enabled output keying information sufficient, server
10 side, to decode a TLS 1.3 packet capture.
12 JH/02 OpenSSL: Suppress the sending of (stateful) TLS1.3 session tickets.
13 Previously the default library behaviour applied, sending two, each in
16 JH/03 Debug output for ACL now gives the config file name and line number for
19 JH/04 The default received_header_text now uses the RFC 8314 tls cipher clause.
21 JH/05 DKIM: ensure that dkim_domain elements are lowercased before use.
23 JH/06 Fix buggy handling of autoreply bounce_return_size_limit, and a possible
24 buffer overrun for (non-chunking) other transports.
26 JH/07 GnuTLS: Our use of late (post-handshake) certificate verification, under
27 TLS1.3, means that a server rejecting a client certificate is not visible
28 to the client until the first read of encrypted data (typically the
29 response to EHLO). Add detection for that case and treat it as a failed
30 TLS connection attempt, so that the normal retry-in-clear can work (if
33 JB/01 Bug 2375: fix expansions of 822 addresses having comments in local-part
34 and/or domain. Found and fixed by Jason Betts.
36 JH/08 Add hardening against SRV & TLSA lookups the hit CNAMEs (a nonvalid
37 configuration). If a CNAME target was not a wellformed name pattern, a
40 JH/09 Logging: Fix initial listening-on line for multiple ports for an IP when
41 the OS reports them interleaved with other addresses.
43 JH/10 OpenSSL: Fix aggregation of messages. Previously, when PIPELINING was
44 used both for input and for a verify callout, both encrypted, SMTP
45 responses being sent by the server could be lost. This resulted in
46 dropped connections and sometimes bounces generated by a peer sending
49 JH/11 Harden plaintext authenticator against a badly misconfigured client-send
50 string. Previously it was possible to cause undefined behaviour in a
51 library routine (usually a crash). Found by "zerons".
53 JH/12 Bug 2384: fix "-bP smtp_receive_timeout". Previously it returned no
56 JH/13 Bug 2386: Fix builds with Dane under LibreSSL 2.9.0 onward. Some old
57 API was removed, so update to use the newer ones.
59 JH/14 Bug 1891: Close the log file if receiving a non-smtp message, without
60 any timeout set, is taking a long time. Previously we would hang on to a
61 rotated logfile "forever" if the input was arriving with long gaps
62 (a previous attempt to fix addressed lack, for a long time, of initial
65 HS/01 Bug 2390: Use message_id for tempfile creation to avoid races in a
66 shared (NFS) environment. The length of the tempfile name is now
67 4 + 16 ("hdr.$message_exim_id") which might break on file
68 systems which restrict the file name length to lower values.
71 HS/02 Bug 2390: Use message_id for tempfile creation to avoid races in a
72 shared (NFS) environment.
74 HS/03 Bug 2392: exigrep does case sensitive *option* processing (as it
75 did for all versions <4.90). Notably -M, -m, --invert, -I may be
78 JH/15 Use unsigned when creating bitmasks in macros, to avoid build errors
79 on some platforms for bit 31.
81 JH/16 GnuTLS: rework ciphersuite strings under recent library versions. Thanks
82 to changes apparently associated with TLS1.3 handling some of the APIs
83 previously used were either nonfunctional or inappropriate. Strings
84 like TLS1.3:ECDHE_SECP256R1__RSA_PSS_RSAE_SHA256__AES_256_GCM__AEAD:256
85 and TLS1.2:ECDHE_SECP256R1__RSA_SHA256__AES_128_CBC__SHA256:128 replace
86 the previous TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256 .
87 This affects log line X= elements, the $tls_{in,out}_cipher variables,
88 and the use of specific cipher names in the encrypted= ACL condition.
90 JH/17 OpenSSL: the default openssl_options now disables ssl_v3.
92 JH/18 GnuTLS: fix $tls_out_ocsp under hosts_request_ocsp. Previously the
93 verification result was not updated unless hosts_require_ocsp applied.
95 JH/19 Bug 2398: fix listing of a named-queue. Previously, even with the option
96 queue_list_requires_admin set to false, non-admin users were denied the
99 JH/20 Bug 2389: fix server advertising of usable certificates, under GnuTLS in
100 directory-of-certs mode. Previously they were advertised despite the
103 JH/21 The smtp transport option "hosts_noproxy_tls" is now unset by default.
104 A single TCP connection by a client will now hold a TLS connection open
105 for multiple message deliveries, by default. Previoud the default was to
108 JH/22 The smtp transport option "hosts_try_dane" now enables all hosts by
109 default. If built with the facility, DANE will be used. The facility
110 SUPPORT_DANE is now enabled in the prototype build Makefile "EDITME".
112 JH/23 The build default is now for TLS to be included; the SUPPORT_TLS define
113 is replaced with DISABLE_TLS. Either USE_GNUTLS or (the new) USE_OPENSSL
114 must be defined and you must still, unless you define DISABLE_TLS, manage
115 the the include-dir and library-file requirements that go with that
116 choice. Non-TLS builds are still supported.
118 JH/24 Fix duplicated logging of peer name/address, on a transport connection-
121 JH/25 The smtp transport option "hosts_try_fastopen" now enables all hosts by
122 default. If the platform supports and has the facility enabled, it will
123 be requested on all coneections.
125 JH/26 The PIPE_CONNECT facility is promoted from experimental status and is now
126 controlled by the build-time option SUPPORT_PIPE_CONNECT.
128 PP/01 Unbreak heimdal_gssapi, broken in 4.92.
130 JH/27 Bug 2404: Use the main-section configuration option "dsn_from" for
131 success-DSN messages. Previously the From: header was always the default
132 one for these; the option was ignored.
134 JH/28 Fix the timeout on smtp response to apply to the whole response.
135 Previously it was reset for every read, so a teergrubing peer sending
136 single bytes within the time limit could extend the connection for a
137 long time. Credit to Qualsys Security Advisory Team for the discovery.
139 JH/29 Fix DSN Final-Recipient: field. Previously it was the post-routing
140 delivery address, which leaked information of the results of local
141 forwarding. Change to the original envelope recipient address, per
144 JH/30 Bug 2411: Fix DSN generation when RFC 3461 failure notification is
145 requested. Previously not bounce was generated and a log entry of
146 error ignored was made.
148 JH/31 Avoid re-expansion in ${sort } expansion. (CVE-2019-13917)
150 JH/32 Introduce a general tainting mechanism for values read from the input
151 channel, and values derived from them. Refuse to expand any tainted
152 values, to catch one form of exploit.
154 JH/33 Bug 2413: Fix dkim_strict option. Previously the expansion result
155 was unused and the unexpanded text used for the test. Found and
156 fixed by Ruben Jenster.
158 JH/34 Fix crash after TLS shutdown. When the TCP/SMTP channel was left open,
159 an attempt to use a TLS library read routine dereffed a nul pointer,
162 JH/35 Bug 2409: filter out-of-spec chars from callout response before using
163 them in our smtp response.
165 JH/36 Have the general router option retry_use_local_part default to true when
166 any of the restrictive preconditions are set (to anything). Previously it
167 was only for check_local user. The change removes one item of manual
168 configuration which is required for proper retries when a remote router
169 handles a subset of addresses for a domain.
171 JH/37 Appendfile: when evaluating quota use (non-quota_size_regex) take the file
172 link count into consideration.
174 HS/04 Fix handling of very log lines in -H files. If a -<key> <value> line
175 caused the extension of big_buffer, the following lines were ignored.
177 JH/38 Bug 1395: Teach the DNS negative-cache about TTL value from the SOA in
178 accordance with RFC 2308. Previously there was no expiry, so a longlived
179 receive process (eg. due to ACL delays) versus a short SOA value could
182 HS/05 Handle trailing backslash gracefully. (CVE-2019-15846)
184 JH/39 Promote DMARC support to mainline.
186 JH/40 Bug 2452: Add a References: header to DSNs.
188 JH/41 With GnuTLS 3.6.0 (and later) do not attempt to manage Diffie-Hellman
189 parameters. The relevant library call is documented as "Deprecated: This
190 function is unnecessary and discouraged on GnuTLS 3.6.0 or later. Since
191 3.6.0, DH parameters are negotiated following RFC7919."
193 HS/06 Change the default of dnssec_request_domains to "*"
198 JH/01 Remove code calling the customisable local_scan function, unless a new
199 definition "HAVE_LOCAL_SCAN=yes" is present in the Local/Makefile.
201 JH/02 Bug 1007: Avoid doing logging from signal-handlers, as that can result in
202 non-signal-safe functions being used.
204 JH/03 Bug 2269: When presented with a received message having a stupidly large
205 number of DKIM-Signature headers, disable DKIM verification to avoid
206 a resource-consumption attack. The limit is set at twenty.
208 JH/04 Add variables $arc_domains, $arc_oldest_pass for ARC verify. Fix the
209 report of oldest_pass in ${authres } in consequence, and separate out
210 some descriptions of reasons for verification fail.
212 JH/05 Bug 2273: Cutthrough delivery left a window where the received messsage
213 files in the spool were present and unlocked. A queue-runner could spot
214 them, resulting in a duplicate delivery. Fix that by doing the unlock
215 after the unlink. Investigation by Tim Stewart. Take the opportunity to
216 add more error-checking on spoolfile handling while that code is being
219 PP/01 Refuse to open a spool data file (*-D) if it's a symlink.
220 No known attacks, no CVE, this is defensive hardening.
222 JH/06 Bug 2275: The MIME ACL unlocked the received message files early, and
223 a queue-runner could start a delivery while other operations were ongoing.
224 Cutthrough delivery was a common victim, resulting in duplicate delivery.
225 Found and investigated by Tim Stewart. Fix by using the open message data
226 file handle rather than opening another, and not locally closing it (which
227 releases a lock) for that case, while creating the temporary .eml format
228 file for the MIME ACL. Also applies to "regex" and "spam" ACL conditions.
230 JH/07 Bug 177: Make a random-recipient callout success visible in ACL, by setting
231 $sender_verify_failure/$recipient_verify_failure to "random".
233 JH/08 When generating a selfsigned cert, use serial number 1 since zero is not
236 JH/09 Bug 2274: Fix logging of cmdline args when starting in an unlinked cwd.
237 Previously this would segfault.
239 JH/10 Fix ARC signing for case when DKIM signing failed. Previously this would
242 JH/11 Bug 2264: Exim now only follows CNAME chains one step by default. We'd
243 like zero, since the resolver should be doing this for us, But we need one
244 as a CNAME but no MX presence gets the CNAME returned; we need to check
245 that doesn't point to an MX to declare it "no MX returned" rather than
246 "error, loop". A new main option is added so the older capability of
247 following some limited number of chain links is maintained.
249 JH/12 Add client-ip info to non-pass iprev ${authres } lines.
251 JH/13 For receent Openssl versions (1.1 onward) use modern generic protocol
252 methods. These should support TLS 1.3; they arrived with TLS 1.3 and the
253 now-deprecated earlier definitions used only specified the range up to TLS
254 1.2 (in the older-version library docs).
256 JH/14 Bug 2284: Fix DKIM signing for body lines starting with a pair of dots.
258 JH/15 Rework TLS client-side context management. Stop using a global, and
259 explicitly pass a context around. This enables future use of TLS for
260 connections to service-daemons (eg. malware scanning) while a client smtp
261 connection is using TLS; with cutthrough connections this is quite likely.
263 JH/16 Fix ARC verification to do AS checks in reverse order.
265 JH/17 Support a "tls" option on the ${readsocket } expansion item.
267 JH/18 Bug 2287: Fix the protocol name (eg utf8esmtp) for multiple messages
268 using the SMTPUTF8 option on their MAIL FROM commands, in one connection.
269 Previously the "utf8" would be re-prepended for every additional message.
271 JH/19 Reject MAIL FROM commands with SMTPUTF8 when the facility was not advertised.
272 Previously thery were accepted, resulting in issues when attempting to
273 forward messages to a non-supporting MTA.
275 PP/02 Let -n work with printing macros too, not just options.
277 JH/20 Bug 2296: Fix cutthrough for >1 address redirection. Previously only
278 one parent address was copied, and bogus data was used at delivery-logging
279 time. Either a crash (after delivery) or bogus log data could result.
280 Discovery and analysis by Tim Stewart.
282 PP/03 Make ${utf8clean:} expansion operator detect incomplete final character.
283 Previously if the string ended mid-character, we did not insert the
284 promised '?' replacement.
286 PP/04 Documentation: current string operators work on bytes, not codepoints.
288 JH/21 Change as many as possible of the global flags into one-bit bitfields; these
289 should pack well giving a smaller memory footprint so better caching and
290 therefore performance. Group the declarations where this can't be done so
291 that the byte-sized flag variables are not interspersed among pointer
292 variables, giving a better chance of good packing by the compiler.
294 JH/22 Bug 1896: Fix the envelope from for DMARC forensic reports to be possibly
295 non-null, to avoid issues with sites running BATV. Previously reports were
296 sent with an empty envelope sender so looked like bounces.
298 JH/23 Bug 2318: Fix the noerror command within filters. It wasn't working.
299 The ignore_error flag wasn't being returned from the filter subprocess so
300 was not set for later routers. Investigation and fix by Matthias Kurz.
302 JH/24 Bug 2310: Raise a msg:fail:internal event for each undelivered recipient,
303 and a msg:complete for the whole, when a message is manually removed using
304 -Mrm. Developement by Matthias Kurz, hacked on by JH.
306 JH/25 Avoid fixed-size buffers for pathnames in DB access. This required using
307 a "Gnu special" function, asprintf() in the DB utility binary builds; I
308 hope that is portable enough.
310 JH/26 Bug 2311: Fix DANE-TA verification under GnuTLS. Previously it was also
311 requiring a known-CA anchor certificate; make it now rely entirely on the
312 TLSA as an anchor. Checking the name on the leaf cert against the name
313 on the A-record for the host is still done for TA (but not for EE mode).
315 JH/27 Fix logging of proxy address. Previously, a pointless "PRX=[]:0" would be
316 included in delivery lines for non-proxied connections, when compiled with
317 SUPPORT_SOCKS and running with proxy logging enabled.
319 JH/28 Bug 2314: Fire msg:fail:delivery event even when error is being ignored.
320 Developement by Matthias Kurz, tweaked by JH. While in that bit of code,
321 move the existing event to fire before the normal logging of message
322 failure so that custom logging is bracketed by normal logging.
324 JH/29 Bug 2322: A "fail" command in a non-system filter (file) now fires the
325 msg:fail:internal event. Developement by Matthias Kurz.
327 JH/30 Bug 2329: Increase buffer size used for dns lookup from 2k, which was
328 far too small for todays use of crypto signatures stored there. Go all
329 the way to the max DNS message size of 64kB, even though this might be
330 overmuch for IOT constrained device use.
332 JH/31 Fix a bad use of a copy function, which could be used to pointlessly
333 copy a string over itself. The library routine is documented as not
334 supporting overlapping copies, and on MacOS it actually raised a SIGABRT.
336 JH/32 For main options check_spool_space and check_inode_space, where the
337 platform supports 64b integers, support more than the previous 2^31 kB
338 (i.e. more than 2 TB). Accept E, P and T multipliers in addition to
339 the previous G, M, k.
341 JH/33 Bug 2338: Fix the cyrus-sasl authenticator to fill in the
342 $authenticated_fail_id variable on authentication failure. Previously
345 JH/34 Increase RSA keysize of autogen selfsign cert from 1024 to 2048. RHEL 8.0
346 OpenSSL didn't want to use such a weak key. Do for GnuTLS also, and for
347 more-modern GnuTLS move from GNUTLS_SEC_PARAM_LOW to
348 GNUTLS_SEC_PARAM_MEDIUM.
350 JH/35 OpenSSL: fail the handshake when SNI processing hits a problem, server
351 side. Previously we would continue as if no SNI had been received.
353 JH/36 Harden the handling of string-lists. When a list consisted of a sole
354 "<" character, which should be a list-separator specification, we walked
355 off past the nul-terimation.
357 JH/37 Bug 2341: Send "message delayed" warning MDNs (restricted to external
358 causes) even when the retry time is not yet met. Previously they were
359 not, meaning that when (say) an account was over-quota and temp-rejecting,
360 and multiple senders' messages were queued, only one sender would get
361 notified on each configured delay_warning cycle.
363 JH/38 Bug 2351: Log failures to extract envelope addresses from message headers.
365 JH/39 OpenSSL: clear the error stack after an SSL_accept(). With anon-auth
366 cipher-suites, an error can be left on the stack even for a succeeding
367 accept; this results in impossible error messages when a later operation
370 AM/01 Bug 2359: GnuTLS: repeat lowlevel read and write operations while they
371 return error codes indicating retry. Under TLS1.3 this becomes required.
373 JH/40 Fix the feature-cache refresh for EXPERIMENTAL_PIPE_CONNECT. Previously
374 it only wrote the new authenticators, resulting in a lack of tracking of
375 peer changes of ESMTP extensions until the next cache flush.
377 JH/41 Fix the loop reading a message header line to check for integer overflow,
378 and more-often against header_maxsize. Previously a crafted message could
379 induce a crash of the recive process; now the message is cleanly rejected.
381 JH/42 Bug 2366: Fix the behaviour of the dkim_verify_signers option. It had
382 been totally disabled for all of 4.91. Discovery and fix by "Mad Alex".
388 GF/01 DEFER rather than ERROR on redis cluster MOVED response.
389 When redis_servers is set to a list of > 1 element, and the Redis servers
390 in that list are in cluster configuration, convert the REDIS_REPLY_ERROR
391 case of MOVED into a DEFER case instead, thus moving the query onto the
392 next server in the list. For a cluster of N elements, all N servers must
393 be defined in redis_servers.
395 GF/02 Catch and remove uninitialized value warning in exiqsumm
396 Check for existence of @ARGV before looking at $ARGV[0]
398 JH/01 Replace the store_release() internal interface with store_newblock(),
399 which internalises the check required to safely use the old one, plus
400 the allocate and data copy operations duplicated in both (!) of the
401 extant use locations.
403 JH/02 Disallow '/' characters in queue names specified for the "queue=" ACL
404 modifier. This matches the restriction on the commandline.
406 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
407 Previously only the last row was returned.
409 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
410 we assumed that tags in the header were well-formed, and parsed the
411 element content after inspecting only the first char of the tag.
412 Assumptions at that stage could crash the receive process on malformed
415 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
416 While running the DKIM ACL we operate on the Permanent memory pool so that
417 variables created with "set" persist to the DATA ACL. Also (at any time)
418 DNS lookups that fail create cache records using the Permanent pool. But
419 expansions release any allocations made on the current pool - so a dnsdb
420 lookup expansion done in the DKIM ACL releases the memory used for the
421 DNS negative-cache, and bad things result. Solution is to switch to the
422 Main pool for expansions.
423 While we're in that code, add checks on the DNS cache during store_reset,
424 active in the testsuite.
425 Problem spotted, and debugging aided, by Wolfgang Breyha.
427 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
428 When none of the hosts presented to a transport match an already-open
429 connection, close it and proceed with the list. Previously we would
430 queue the message. Spotted by Lena with Yahoo, probably involving
433 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
434 Previously a spurious "250 OK id=" response was appended to the proper
437 JH/08 The "support for" informational output now, which built with Content
438 Scanning support, has a line for the malware scanner interfaces compiled
439 in. Interface can be individually included or not at build time.
441 JH/09 The "aveserver", "kavdaemon" and "mksd" interfaces are now not included
442 by the template makefile "src/EDITME". The "STREAM" support for an older
443 ClamAV interface method is removed.
445 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
446 rows affected is given instead).
448 JH/11 The runtime Berkeley DB library version is now additionally output by
449 "exim -d -bV". Previously only the compile-time version was shown.
451 JH/12 Bug 2230: Fix cutthrough routing for nonfirst messages in an initiating
452 SMTP connection. Previously, when one had more receipients than the
453 first, an abortive onward connection was made. Move to full support for
454 multiple onward connections in sequence, handling cutthrough connection
455 for all multi-message initiating connections.
457 JH/13 Bug 2229: Fix cutthrough routing for nonstandard port numbers defined by
458 routers. Previously, a multi-recipient message would fail to match the
459 onward-connection opened for the first recipient, and cause its closure.
461 JH/14 Bug 2174: A timeout on connect for a callout was also erroneously seen as
462 a timeout on read on a GnuTLS initiating connection, resulting in the
463 initiating connection being dropped. This mattered most when the callout
464 was marked defer_ok. Fix to keep the two timeout-detection methods
467 JH/15 Relax results from ACL control request to enable cutthrough, in
468 unsupported situations, from error to silently (except under debug)
469 ignoring. This covers use with PRDR, frozen messages, queue-only and
472 HS/01 Fix Buffer overflow in base64d() (CVE-2018-6789)
474 JH/16 Fix bug in DKIM verify: a buffer overflow could corrupt the malloc
475 metadata, resulting in a crash in free().
477 PP/01 Fix broken Heimdal GSSAPI authenticator integration.
478 Broken in f2ed27cf5, missing an equals sign for specified-initialisers.
479 Broken also in d185889f4, with init system revamp.
481 JH/17 Bug 2113: Fix conversation closedown with the Avast malware scanner.
482 Previously we abruptly closed the connection after reading a malware-
483 found indication; now we go on to read the "scan ok" response line,
486 JH/18 Bug 2239: Enforce non-usability of control=utf8_downconvert in the mail
487 ACL. Previously, a crash would result.
489 JH/19 Speed up macro lookups during configuration file read, by skipping non-
490 macro text after a replacement (previously it was only once per line) and
491 by skipping builtin macros when searching for an uppercase lead character.
493 JH/20 DANE support moved from Experimental to mainline. The Makefile control
494 for the build is renamed.
496 JH/21 Fix memory leak during multi-message connections using STARTTLS. A buffer
497 was allocated for every new TLS startup, meaning one per message. Fix
498 by only allocating once (OpenSSL) or freeing on TLS-close (GnuTLS).
500 JH/22 Bug 2236: When a DKIM verification result is overridden by ACL, DMARC
501 reported the original. Fix to report (as far as possible) the ACL
502 result replacing the original.
504 JH/23 Fix memory leak during multi-message connections using STARTTLS under
505 OpenSSL. Certificate information is loaded for every new TLS startup,
506 and the resources needed to be freed.
508 JH/24 Bug 2242: Fix exim_dbmbuild to permit directoryless filenames.
510 JH/25 Fix utf8_downconvert propagation through a redirect router. Previously it
513 JH/26 Bug 2253: For logging delivery lines under PRDR, append the overall
514 DATA response info to the (existing) per-recipient response info for
515 the "C=" log element. It can have useful tracking info from the
516 destination system. Patch from Simon Arlott.
518 JH/27 Bug 2251: Fix ldap lookups that return a single attribute having zero-
519 length value. Previously this would segfault.
521 HS/02 Support Avast multiline protoocol, this allows passing flags to
522 newer versions of the scanner.
524 JH/28 Ensure that variables possibly set during message acceptance are marked
525 dead before release of memory in the daemon loop. This stops complaints
526 about them when the debug_store option is enabled. Discovered specifically
527 for sender_rate_period, but applies to a whole set of variables.
528 Do the same for the queue-runner and queue-list loops, for variables set
529 from spool message files. Do the same for the SMTP per-message loop, for
530 certain variables indirectly set in ACL operations.
532 JH/29 Bug 2250: Fix a longstanding bug in heavily-pipelined SMTP input (such
533 as a multi-recipient message from a mailinglist manager). The coding had
534 an arbitrary cutoff number of characters while checking for more input;
535 enforced by writing a NUL into the buffer. This corrupted long / fast
536 input. The problem was exposed more widely when more pipelineing of SMTP
537 responses was introduced, and one Exim system was feeding another.
538 The symptom is log complaints of SMTP syntax error (NUL chars) on the
539 receiving system, and refused recipients seen by the sending system
540 (propating to people being dropped from mailing lists).
541 Discovered and pinpointed by David Carter.
543 JH/30 The (EXPERIMENTAL_DMARC) variable $dmarc_ar_header is withdrawn, being
544 replaced by the ${authresults } expansion.
546 JH/31 Bug 2257: Fix pipe transport to not use a socket-only syscall.
548 HS/03 Set a handler for SIGTERM and call exit(3) if running as PID 1. This
549 allows proper process termination in container environments.
551 JH/32 Bug 2258: Fix spool_wireformat in combination with LMTP transport.
552 Previously the "final dot" had a newline after it; ensure it is CR,LF.
554 JH/33 SPF: remove support for the "spf" ACL condition outcome values "err_temp"
555 and "err_perm", deprecated since 4.83 when the RFC-defined words
556 "temperror" and "permerror" were introduced.
558 JH/34 Re-introduce enforcement of no cutthrough delivery on transports having
559 transport-filters or DKIM-signing. The restriction was lost in the
560 consolidation of verify-callout and delivery SMTP handling.
561 Extend the restriction to also cover ARC-signing.
563 JH/35 Cutthrough: for a final-dot response timeout (and nonunderstood responses)
564 in defer=pass mode supply a 450 to the initiator. Previously the message
567 PP/02 DANE: add dane_require_tls_ciphers SMTP Transport option; if unset,
568 tls_require_ciphers is used as before.
570 HS/03 Malware Avast: Better match the Avast multiline protocol. Add
571 "pass_unscanned". Only tmpfails from the scanner are written to
572 the paniclog, as they may require admin intervention (permission
573 denied, license issues). Other scanner errors (like decompression
574 bombs) do not cause a paniclog entry.
576 JH/36 Fix reinitialisation of DKIM logging variable between messages.
577 Previously it was possible to log spurious information in receive log
580 JH/37 Bug 2255: Revert the disable of the OpenSSL session caching. This
581 triggered odd behaviour from Outlook Express clients.
583 PP/03 Add util/renew-opendmarc-tlds.sh script for safe renewal of public
586 JH/38 DKIM: accept Ed25519 pubkeys in SubjectPublicKeyInfo-wrapped form,
587 since the IETF WG has not yet settled on that versus the original
588 "bare" representation.
590 JH/39 Fix syslog logging for syslog_timestamp=no and log_selector +millisec.
591 Previously the millisecond value corrupted the output.
592 Fix also for syslog_pid=no and log_selector +pid, for which the pid
593 corrupted the output.
599 JH/01 Rework error string handling in TLS interface so that the caller in
600 more cases is responsible for logging. This permits library-sourced
601 string to be attached to addresses during delivery, and collapses
602 pairs of long lines into single ones.
604 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
605 during configuration. Wildcards are allowed and expanded.
607 JH/02 Rework error string handling in DKIM to pass more info back to callers.
608 This permits better logging.
610 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
611 do not close it down and have the child transport start it up again on
612 the passed-on TCP connection. Instead, proxy the child (and any
613 subsequent ones) for TLS via a unix-domain socket channel. Logging is
614 affected: the continued delivery log lines do not have any DNSSEC, TLS
615 Certificate or OCSP information. TLS cipher information is still logged.
617 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
618 identical IP addresses on different listening ports. Will also affect
621 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
622 add noisy ifdef guards to special-case this sillyness.
623 Patch from Bernd Kuhls.
625 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
626 than 255 are no longer allowed.
628 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
629 Disable the session-cache too, which might reduce our load. Since we
630 currrectly use a new context for every connection, both as server and
631 client, there is no benefit for these.
632 GnuTLS appears to not support tickets server-side by default (we don't
633 call gnutls_session_ticket_enable_server()) but client side is enabled
634 by default on recent versions (3.1.3 +) unless the PFS priority string
637 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
638 <https://reproducible-builds.org/specs/source-date-epoch/>.
640 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
641 the check for any unsuccessful recipients did not notice the limit, and
642 erroneously found still-pending ones.
644 JH/08 Pipeline CHUNKING command and data together, on kernels that support
645 MSG_MORE. Only in-clear (not on TLS connections).
647 JH/09 Avoid using a temporary file during transport using dkim. Unless a
648 transport-filter is involved we can buffer the headers in memory for
649 creating the signature, and read the spool data file once for the
650 signature and again for transmission.
652 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
653 4.77 as the kernel support then wasn't solid, having issues in 64bit
654 mode. Now, it's been long enough. Add support for FreeBSD also.
656 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
657 case where the routing stage had gathered several addresses to send to
658 a host before calling the transport for the first, we previously failed
659 to close down TLS in the old transport process before passing the TCP
660 connection to the new process. The new one sent a STARTTLS command
661 which naturally failed, giving a failed delivery and bloating the retry
662 database. Investigation and fix prototype from Wolfgang Breyha.
664 JH/12 Fix check on SMTP command input synchronisation. Previously there were
665 false-negatives in the check that the sender had not preempted a response
666 or prompt from Exim (running as a server), due to that code's lack of
667 awareness of the SMTP input buffering.
669 PP/04 Add commandline_checks_require_admin option.
670 Exim drops privileges sanely, various checks such as -be aren't a
671 security problem, as long as you trust local users with access to their
672 own account. When invoked by services which pass untrusted data to
673 Exim, this might be an issue. Set this option in main configuration
674 AND make fixes to the calling application, such as using `--` to stop
677 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
678 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
679 responses to those, into a single TLS record each way (this usually means
680 a single packet). As a side issue, smtp_enforce_sync now works on TLS
683 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
684 affects you only if you're dancing at the edge of the param size limits.
685 If you are, and this message makes sense to you, then: raise the
686 configured limit or use OpenSSL 1.1. Nothing we can do for older
689 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
690 cmdline element to get the documented default one. Previously it was
693 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
694 are unsuable for non-message-specific reasons - eg. connection timeout,
695 banner-time rejection.
697 JH/16 Fix logging of delivery remote port, when specified by router, under
700 PP/06 Repair manualroute's ability to take options in any order, even if one
701 is the name of a transport.
704 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
706 JH/17 Change the list-building routines interface to use the expanding-string
707 triplet model, for better allocation and copying behaviour.
709 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
710 Previously it was constructed the first time a possibly-matching string
711 was met in the configuration file input during startup; now it is done
714 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
715 compatible one, to avoid the (poorly documented) possibility of a config
716 file in the working directory redirecting the DB files, possibly correpting
717 some existing file. CVE-2017-10140 assigned for BDB.
719 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
720 cache-hot. Previously, although the result was properly cached, the
721 initial verify call returned a defer.
723 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
724 the main verify for receipient in uncached-mode.
726 JH/22 Retire historical build files to an "unsupported" subdir. These are
727 defined as "ones for which we have no current evidence of testing".
729 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
730 if present. Previously it was ignored.
732 JH/24 Start using specified-initialisers in C structure init coding. This is
733 a C99 feature (it's 2017, so now considered safe).
735 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
736 if was a fixed-sized field and bitmask ops via macros; it is now more
739 PP/07 GitHub PR 56: Apply MariaDB build fix.
740 Patch provided by Jaroslav Škarvada.
742 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
743 during Coverity cleanups [4.87 JH/47]
744 Diagnosis and fix provided by Michael Fischer v. Mollard.
746 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
747 the right size to place the terminating semicolon on its own folded
748 line, the header hash was calculated to an incorrect value thanks to
749 the (relaxed) space the fold became.
751 HS/02 Fix Bug 2130: large writes from the transport subprocess were chunked
752 and confused the parent.
754 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
755 which could crash as a result. This could lead to undeliverable messages.
757 JH/28 Logging: "next input sent too soon" now shows where input was truncated
760 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
761 matters on fast-turnover and PID-randomising systems, which were getting
762 out-of-order delivery.
764 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
765 a possibly-overlapping copy. The symptom was that "Remote host closed
766 connection in response to HELO" was logged instead of the actual 4xx
769 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
770 Previously only that bufferd was discarded, resulting in SYMTP command
773 JH/32 DKIM: when a message has multiple signatures matching an identity given
774 in dkim_verify_signers, run the dkim acl once for each. Previously only
775 one run was done. Bug 2189.
777 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
778 "panic the current process" to "deliberately defer". The panic log is
779 still written with the problem list name; the mail and reject logs now
780 get a temp-reject line for the message that was being handled, saying
781 something like "domains check lookup or other defer". The SMTP 451
782 message is still "Temporary local problem".
784 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
785 A crafted sequence of BDAT commands could result in in-use memory beeing
786 freed. CVE-2017-16943.
788 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
789 from SMTP input. Previously it was always done; now only done for DATA
790 and not BDAT commands. CVE-2017-16944.
792 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
793 to the message (such as an overlong header line). Previously this was
794 not done and we did not exit BDAT mode. Followon from the previous item
795 though a different problem.
801 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
802 than -2003 did; needs libidn2 in addition to libidn.
804 JH/02 The path option on a pipe transport is now expanded before use.
806 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
807 Patch provided by "Björn", documentation fix added too.
809 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
810 missing a wire-to-host endian conversion.
812 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
813 close after a BDAT command line could be taken as a following command,
814 giving a synch failure. Fix by only checking for synch immediately
815 before acknowledging the chunk.
817 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
818 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
819 macro. Patches provided by Josh Soref.
821 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
822 Previously we did not; the RFC seems ambiguous and VRFY is not listed
823 by IANA as a service extension. However, John Klensin suggests that we
826 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
827 the dkim code may be unix-mode line endings rather than smtp wire-format
828 CRLF, so prepend a CR to any bare LF.
830 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
831 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
833 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
834 the first were themselves being wrongly included in the feed into dkim
835 processing; with most chunk sizes in use this resulted in an incorrect
836 body hash calculated value.
838 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
839 DKIM signature block, for verification. Although advised against by
840 standards it is specifically not ruled illegal.
842 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
844 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
845 missing a body hash (the bh= tag).
847 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
848 It seems that HAProxy sends the Proxy Protocol information in clear and
849 only then does a TLS startup, so do the same.
851 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
852 TCP connections (such as for Spamd) unless the daemon successfully set
853 Fast Open mode on its listening sockets. This fixes breakage seen on
854 too-old kernels or those not configured for Fast Open, at the cost of
855 requiring both directions being enabled for TFO, and TFO never being used
856 by non-daemon-related Exim processes.
858 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
859 endings, at least on the first header line. Try to canonify any that get
860 past that check, despite the cost.
862 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
863 now limited to an arbitrary five deep, while parsing addresses with the
864 strip_excess_angle_brackets option enabled.
866 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
867 instead leave the unprompted TLS handshake in socket buffer for the
868 TLS library to consume.
870 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
872 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
874 JH/16 Drop variables when they go out of scope. Memory management drops a whole
875 region in one operation, for speed, and this leaves assigned pointers
876 dangling. Add checks run only under the testsuite which checks all
877 variables at a store-reset and panics on a dangling pointer; add code
878 explicitly nulling out all the variables discovered. Fixes one known
879 bug: a transport crash, where a dangling pointer for $sending_ip_address
880 originally assigned in a verify callout, is re-used.
882 PP/06 Drop '.' from @INC in various Perl scripts.
884 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
886 PP/08 Reduce a number of compilation warnings under clang; building with
887 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
888 should be warning-free.
890 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
892 HS/01 Fix portability problems introduced by PP/08 for platforms where
893 realloc(NULL) is not equivalent to malloc() [SunOS et al].
895 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
896 chunk. This allows us to accept broken chunked messages. We need a more
897 general solution here.
899 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
900 already-broken messages in the queue.
902 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
904 JH/19 Fix reference counting bug in routing-generated-address tracking.
910 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
911 supports it and a size is available (ie. the sending peer gave us one).
913 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
914 years of being deprecated). The replacements are the ACLs
915 acl_smtp_mime and acl_not_smtp_mime.
917 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
918 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
919 or even an in-clear connection were permitted. Now, if the host lookup
920 was dnssec and dane was requested then the host is only used if the
921 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
922 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
923 if one fails this test.
924 This means that a poorly-configured remote DNS will make it incommunicado;
925 but it protects against a DNS-interception attack on it.
927 JH/04 Bug 1810: make continued-use of an open smtp transport connection
928 non-noisy when a race steals the message being considered.
930 JH/05 If main configuration option tls_certificate is unset, generate a
931 self-signed certificate for inbound TLS connections.
933 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
934 in rewrites and routers.
936 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
937 and logged a warning sing 4.83; now they are a configuration file error.
939 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
940 (lacking @domain). Apply the same qualification processing as RCPT.
942 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
944 JH/10 Support ${sha256:} applied to a string (as well as the previous
947 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
948 a cutthrough deliver is pending, as we always want to make a connection.
949 This also avoids re-routing the message when later placing the cutthrough
950 connection after a verify cache hit.
951 Do not update it with the verify result either.
953 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
954 when routing results in more than one destination address.
956 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
957 signing (which inhibits the cutthrough capability). Previously only
958 the presence of an option was tested; now an expansion evaluating as
959 empty is permissible (obviously it should depend only on data available
960 when the cutthrough connection is made).
962 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
963 the relevant preceding SMTP command did not note the pipelining mode.
965 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
966 Previously they were not counted.
968 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
969 as one having no matching records. Previously we deferred the message
970 that needed the lookup.
972 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
973 distinguished as "(=".
975 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
976 for missing MX records. Previously it only worked for missing A records.
978 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
980 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
981 after the data-go-ahead and data-ack. Patch from Jason Betts.
983 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
984 even for a "none" policy. Patch from Tony Meyer.
986 JH/22 Fix continued use of a connection for further deliveries. If a port was
987 specified by a router, it must also match for the delivery to be
990 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
991 When one of the recipients of a message was unsuitable for the connection
992 (has no matching addresses), we lost track of needing to mark it
993 deferred. As a result mail would be lost.
995 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
997 JH/25 Decoding ACL controls is now done using a binary search; the source code
998 takes up less space and should be simpler to maintain. Merge the ACL
999 condition decode tables also, with similar effect.
1001 JH/26 Fix problem with one_time used on a redirect router which returned the
1002 parent address unchanged. A retry would see the parent address marked as
1003 delivered, so not attempt the (identical) child. As a result mail would
1006 JH/27 Fix a possible security hole, wherein a process operating with the Exim
1007 UID can gain a root shell. Credit to http://www.halfdog.net/ for
1008 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
1011 JH/28 Enable {spool,log} filesystem space and inode checks as default.
1012 Main config options check_{log,spool}_{inodes,space} are now
1013 100 inodes, 10MB unless set otherwise in the configuration.
1015 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
1016 Previously it only applied to the main-section connection policy
1019 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
1021 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
1022 by me. Added RFC7919 DH primes as an alternative.
1024 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
1025 are not in the system include path.
1027 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
1028 GnuTLS, when a session startup failed (eg because the client disconnected)
1029 Exim did stdio operations after fclose. This was exposed by a recent
1030 change which nulled out the file handle after the fclose.
1032 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
1033 signed directly by the cert-signing cert, rather than an intermediate
1034 OCSP-signing cert. This is the model used by LetsEncrypt.
1036 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
1038 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
1039 an incoming connection.
1041 HS/02 Bug 1802: Do not half-close the connection after sending a request
1044 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
1045 fallback to "prime256v1".
1047 JH/34 SECURITY: Use proper copy of DATA command in error message.
1048 Could leak key material. Remotely exploitable. CVE-2016-9963.
1054 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
1055 and 3.4.4 - once the server is enabled to respond to an OCSP request
1056 it does even when not requested, resulting in a stapling non-aware
1057 client dropping the TLS connection.
1059 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
1060 support variable-length bit vectors. No functional change.
1062 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
1063 The I= interface field on outgoing lines is now after the H= remote
1064 host field, same as incoming lines. There is a separate
1065 outgoing_interface log selector which allows you to disable the
1068 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
1069 If not running log_selector +smtp_connection the mainlog would be held
1070 open indefinitely after a "too many connections" event, including to a
1071 deleted file after a log rotate. Leave the per net connection logging
1072 leaving it open for efficiency as that will be quickly detected by the
1073 check on the next write.
1075 HS/01 Bug 1671: Fix post transport crash.
1076 Processing the wait-<transport> messages could crash the delivery
1077 process if the message IDs didn't exist for some reason. When
1078 using 'split_spool_directory=yes' the construction of the spool
1079 file name failed already, exposing the same netto behaviour.
1081 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
1082 mime_regex ACL conditions.
1084 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
1085 to DSN fail messages (bounces): remote IP, remote greeting, remote response
1086 to HELO, local diagnostic string.
1088 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
1089 log line to debug. Even when configured with a tls authenticator many
1090 client connections are expected to not authenticate in this way, so
1091 an authenticate fail is not an error.
1093 HS/02 Add the Exim version string to the process info. This way exiwhat
1094 gives some more detail about the running daemon.
1096 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
1097 matter for fast-change records such as DNSBLs.
1099 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
1100 retry record, even if constant. There may be multiple transports with
1101 different interface settings and the retry behaviour needs to be kept
1104 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
1106 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
1108 JH/10 Bug 840: fix log_defer_output option of pipe transport
1110 JH/11 Bug 830: use same host for all RCPTS of a message, even under
1111 hosts_randomize. This matters a lot when combined with mua_wrapper.
1113 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
1114 ${quote_pgsql:<string>} operator.
1116 JH/13 Bug 1708: avoid misaligned access in cached lookup.
1118 JH/14 Change header file name for freeradius-client. Relevant if compiling
1119 with Radius support; from the Gentoo tree and checked under Fedora.
1121 JH/15 Bug 1712: Introduce $prdr_requested flag variable
1123 JH/16 Bug 1714: Permit an empty string as expansion result for transport
1124 option transport_filter, meaning no filtering.
1126 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
1128 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
1129 defaults to "*" (all hosts). The variable is now available when not built
1130 with TLS, default unset, mainly to enable keeping the testsuite sane.
1131 If a server certificate is not supplied (via tls_certificate) an error is
1132 logged, and clients will find TLS connections fail on startup. Presumably
1133 they will retry in-clear.
1134 Packagers of Exim are strongly encouraged to create a server certificate
1135 at installation time.
1137 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
1138 with the $config_file variable.
1140 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
1141 in transport context, after the attempt, and per-recipient. The latter type
1142 is per host attempted. The event data is the error message, and the errno
1143 information encodes the lookup type (A vs. MX) used for the (first) host,
1144 and the trailing two digits of the smtp 4xx response.
1146 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
1147 to write to mainlog (or rejectlog, paniclog) in the window between file
1148 creation and permissions/ownership being changed. Particularly affects
1149 installations where exicyclog is run as root, rather than exim user;
1150 result is that the running daemon panics and dies.
1152 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
1154 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
1155 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
1156 "pri" and "weight". Note that the previous implicit priority given by the
1157 list order is no longer honoured.
1159 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
1160 for DKIM processing.
1162 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
1163 by defining SUPPORT_SOCKS.
1165 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
1166 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
1167 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
1168 variables are renamed to proxy_{local,external}_{address,port}.
1170 JH/27 Move Internationalisation support from Experimental to mainline, enabled
1171 for a build by defining SUPPORT_I18N
1173 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
1174 of the query string, and make ${quote_redis:} do that quoting.
1176 JH/29 Move Events support from Experimental to mainline, enabled by default
1177 and removable for a build by defining DISABLE_EVENT.
1179 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
1181 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
1182 cached by the daemon.
1184 JH/32 Move Redis support from Experimental to mainline, enabled for a build
1185 by defining LOOKUP_REDIS. The libhiredis library is required.
1187 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
1188 keys are given for lookup.
1190 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
1191 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
1192 only supported when built with TLS support. The PolarSSL SHA routines
1193 are still used when the TLS library is too old for convenient support.
1195 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
1196 openssl_options), for security. OpenSSL forces this from version 1.1.0
1197 server-side so match that on older versions.
1199 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
1200 allocation for $value could be released as the expansion processing
1201 concluded, but leaving the global pointer active for it.
1203 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
1204 and to use the domains and local_parts ACL conditions.
1206 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
1207 incorrectly not doubled on cutthrough transmission, hence seen as a
1208 body-termination at the receiving system - resulting in truncated mails.
1209 Commonly the sender saw a TCP-level error, and retransmitted the message
1210 via the normal store-and-forward channel. This could result in duplicates
1211 received - but deduplicating mailstores were liable to retain only the
1212 initial truncated version.
1214 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
1216 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
1218 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
1219 we're in there, support oversigning also; bug 1309.
1221 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
1223 HS/04 Add support for keep_environment and add_environment options.
1225 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
1226 either intentional arithmetic overflow during PRNG, or testing config-
1229 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
1230 delivery resulted in actual delivery. Cancel cutthrough before DATA
1233 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
1234 rejects a recipient: pass the reject to the originator.
1236 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
1237 Many were false-positives and ignorable, but it's worth fixing the
1240 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
1241 for the new environment-manipulation done at startup. Move the routines
1242 from being local to tls.c to being global via the os.c file.
1244 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
1245 an extract embedded as result-arg for a map, the first arg for extract
1246 is unavailable so we cannot tell if this is a numbered or keyed
1247 extraction. Accept either.
1253 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
1256 JH/02 The smtp transport option "multi_domain" is now expanded.
1258 JH/03 The smtp transport now requests PRDR by default, if the server offers
1261 JH/04 Certificate name checking on server certificates, when exim is a client,
1262 is now done by default. The transport option tls_verify_cert_hostnames
1263 can be used to disable this per-host. The build option
1264 EXPERIMENTAL_CERTNAMES is withdrawn.
1266 JH/05 The value of the tls_verify_certificates smtp transport and main options
1267 default to the word "system" to access the system default CA bundle.
1268 For GnuTLS, only version 3.0.20 or later.
1270 JH/06 Verification of the server certificate for a TLS connection is now tried
1271 (but not required) by default. The verification status is now logged by
1272 default, for both outbound TLS and client-certificate supplying inbound
1275 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
1278 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
1279 Status Notification (bounce) messages are now MIME format per RFC 3464.
1280 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
1281 under the control of the dsn_advertise_hosts option, and routers may
1282 have a dsn_lasthop option.
1284 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
1285 default, modifiable by a malware= option. The list separator for
1286 the options can now be changed in the usual way. Bug 68.
1288 JH/10 The smtp_receive_timeout main option is now expanded before use.
1290 JH/11 The incoming_interface log option now also enables logging of the
1291 local interface on delivery outgoing connections.
1293 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
1294 if the interface and destination host and port all match.
1296 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
1299 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
1300 Patch from Andrew Lewis.
1302 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
1303 now supports optional time-restrictions, weighting, and priority
1304 modifiers per server. Patch originally by <rommer@active.by>.
1306 JH/16 The spamd_address main option now supports a mixed list of local
1307 and remote servers. Remote servers can be IPv6 addresses, and
1308 specify a port-range.
1310 JH/17 Bug 68: The spamd_address main option now supports an optional
1311 timeout value per server.
1313 JH/18 Bug 1581: Router and transport options headers_add/remove can
1314 now have the list separator specified.
1316 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
1319 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
1322 JH/21 Support for the A6 type of dns record is withdrawn.
1324 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
1325 rather than the verbs used.
1327 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
1328 from 255 to 1024 chars.
1330 JH/24 Verification callouts now attempt to use TLS by default.
1332 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
1333 are generic router options now. The defaults didn't change.
1335 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
1336 Original patch from Alexander Shikoff, worked over by JH.
1338 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
1339 files if ZCAT_COMMAND is not executable.
1341 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
1343 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
1345 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
1346 Normally benign, it bites when the pair was led to by a CNAME;
1347 modern usage is to not canonicalize the domain to a CNAME target
1348 (and we were inconsistent anyway for A-only vs AAAA+A).
1350 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
1352 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
1353 when evaluating $sender_host_dnssec.
1355 JH/31 Check the HELO verification lookup for DNSSEC, adding new
1356 $sender_helo_dnssec variable.
1358 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
1360 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
1362 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
1364 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
1365 documented as working, but never had. Support all but $spam_report.
1367 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
1368 added for tls authenticator.
1370 HS/03 Add perl_taintmode main config option
1376 TL/01 When running the test suite, the README says that variables such as
1377 no_msglog_check are global and can be placed anywhere in a specific
1378 test's script, however it was observed that placement needed to be near
1379 the beginning for it to behave that way. Changed the runtest perl
1380 script to read through the entire script once to detect and set these
1381 variables, reset to the beginning of the script, and then run through
1382 the script parsing/test process like normal.
1384 TL/02 The BSD's have an arc4random API. One of the functions to induce
1385 adding randomness was arc4random_stir(), but it has been removed in
1386 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
1387 function when detected.
1389 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
1390 cause callback expansion.
1392 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
1393 syntax errors in an expansion can be treated as a string instead of
1394 logging or causing an error, due to the internal use of bool_lax
1395 instead of bool when processing it.
1397 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
1398 server certificates when making smtp deliveries.
1400 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
1402 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
1404 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
1406 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
1407 Merged patch from Sebastian Wiedenroth.
1409 JH/05 Fix results-pipe from transport process. Several recipients, combined
1410 with certificate use, exposed issues where response data items split
1411 over buffer boundaries were not parsed properly. This eventually
1412 resulted in duplicates being sent. This issue only became common enough
1413 to notice due to the introduction of connection certificate information,
1414 the item size being so much larger. Found and fixed by Wolfgang Breyha.
1416 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
1417 size buffer was used, resulting in syntax errors when an expansion
1420 JH/07 Add support for directories of certificates when compiled with a GnuTLS
1421 version 3.3.6 or later.
1423 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
1424 is EXPERIMENTAL_EVENT, the main-configuration and transport options
1425 both become "event_action", the variables become $event_name, $event_data
1426 and $event_defer_errno. There is a new variable $verify_mode, usable in
1427 routers, transports and related events. The tls:cert event is now also
1428 raised for inbound connections, if the main configuration event_action
1431 TL/06 In test suite, disable OCSP for old versions of openssl which contained
1432 early OCSP support, but no stapling (appears to be less than 1.0.0).
1434 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
1435 server certificate names available under the smtp transport option
1436 "tls_verify_cert_hostname" now do not permit multi-component wildcard
1439 JH/10 Time-related extraction expansions from certificates now use the main
1440 option "timezone" setting for output formatting, and are consistent
1441 between OpenSSL and GnuTLS compilations. Bug 1541.
1443 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
1444 encoded parameter in the incoming message. Bug 1558.
1446 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
1447 include certificate info, eximon was claiming there were spoolfile
1450 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
1452 JH/14 Log delivery-related information more consistently, using the sequence
1453 "H=<name> [<ip>]" wherever possible.
1455 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
1456 are problematic for Debian distribution, omit them from the release
1459 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
1461 JH/16 Fix string representation of time values on 64bit time_t architectures.
1464 JH/17 Fix a null-indirection in certextract expansions when a nondefault
1465 output list separator was used.
1470 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
1471 checkers that were complaining about end of non-void function with no
1474 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
1475 This was a regression introduced in 4.83 by another bugfix.
1477 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1479 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1480 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1486 TF/01 Correctly close the server side of TLS when forking for delivery.
1488 When a message was received over SMTP with TLS, Exim failed to clear up
1489 the incoming connection properly after forking off the child process to
1490 deliver the message. In some situations the subsequent outgoing
1491 delivery connection happened to have the same fd number as the incoming
1492 connection previously had. Exim would try to use TLS and fail, logging
1493 a "Bad file descriptor" error.
1495 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1496 utilities have not been installed.
1498 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1499 temporary space as the ACL may create new global variables.
1501 TL/01 LDAP support uses per connection or global context settings, depending
1502 upon the detected version of the libraries at build time.
1504 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1505 to extract and use the src ip:port in logging and expansions as if it
1506 were a direct connection from the outside internet. PPv2 support was
1507 updated based on HAProxy spec change in May 2014.
1509 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1511 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1512 Properly escape header and check for NULL return.
1514 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1517 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1519 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1520 characters in header names, implemented as a verify condition.
1521 Contributed by Michael Fischer v. Mollard.
1523 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1524 results permerror and temperror. Previous values are deprecated but
1525 still accepted. In a future release, err_perm and err_temp will be
1526 completely removed, which will be a backward incompatibility if the
1527 ACL tests for either of these two old results. Patch contributed by
1528 user bes-internal on the mailing list.
1530 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1532 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1533 selectors, in both main and reject logs.
1535 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1538 JH/07 Add malware type "sock" for talking to simple daemon.
1540 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1542 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1543 routers/transports under cutthrough routing.
1545 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1546 numbers. Touch up "bool" conditional to keep the same definition.
1548 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1550 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1552 JH/12 Expand items in router/transport headers_add or headers_remove lists
1553 individually rather than the list as a whole. Bug 1452.
1555 Required for reasonable handling of multiple headers_ options when
1556 they may be empty; requires that headers_remove items with embedded
1557 colons must have them doubled (or the list-separator changed).
1559 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1560 view the policy declared in the DMARC record. Currently, $dmarc_status
1561 is a combined value of both the record presence and the result of the
1564 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1566 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1567 dnslookup router and the smtp transport (applying to the forward
1570 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1571 of ldap servers used for a specific lookup. Patch provided by Heiko
1574 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1575 New variable $lookup_dnssec_authenticated for observability.
1577 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1578 Patch submitted by Lars Timman.
1580 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1582 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1583 Requires trusted mode and valid format message id, aborts otherwise.
1584 Patch contributed by Heiko Schlichting.
1586 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1587 certextract with support for various fields. Bug 1358.
1589 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1590 is requested by default, modifiable by smtp transport option
1593 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1594 operate on certificate variables to give certificate fingerprints
1595 Also new ${sha256:cert_variable}.
1597 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1599 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1600 Christian Aistleitner.
1602 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1604 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1605 file. Patch from Wolfgang Breyha.
1607 JH/25 Expand the coverage of the delivery $host and $host_address to
1608 client authenticators run in verify callout. Bug 1476.
1610 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1611 align with daemon_smtp_ports. Bug 72.
1613 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1614 support and error reporting did not work properly.
1616 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1617 and is readable. Patch from Andrew Colin Kissa.
1619 TL/14 Enhance documentation of ${run expansion and how it parses the
1620 commandline after expansion, particularly in the case when an
1621 unquoted variable expansion results in an empty value.
1623 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1625 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1626 by renaming away from it. Addresses GH issue 15, reported by
1629 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1631 TL/15 SECURITY: prevent double expansion in math comparison functions
1632 (can expand unsanitized data). Not remotely exploitable.
1639 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1641 PP/02 Make -n do something, by making it not do something.
1642 When combined with -bP, the name of an option is not output.
1644 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1647 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1648 $sender_host_name and config options to manage this, and basic check
1651 PP/05 DSCP support for outbound connections and control modifier for inbound.
1653 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1654 (Only plugin which currently uses this is kerberos4, which nobody should
1655 be using, but we should make it available and other future plugins might
1656 conceivably use it, even though it would break NAT; stuff *should* be
1657 using channel bindings instead).
1659 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1660 name; added for Sendmail compatibility; requires admin caller.
1661 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1662 just ignore it); requires trusted caller.
1663 Also parse but ignore: -Ac -Am -X<logfile>
1666 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1668 TL/02 Add +smtp_confirmation as a default logging option.
1670 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1671 Patch by Magnus Holmgren from 2007-02-20.
1673 TL/04 Bugzilla 1281 - Spec typo.
1674 Bugzilla 1283 - Spec typo.
1675 Bugzilla 1290 - Spec grammar fixes.
1677 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1679 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1681 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1682 mailing list by by Dmitry Isaikin.
1684 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1686 JH/02 Support "G" suffix to numbers in ${if comparisons.
1688 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1690 NM/01 Bugzilla 1197 - Spec typo
1691 Bugzilla 1196 - Spec examples corrections
1693 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1695 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1696 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1699 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1700 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1702 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1703 "acl {{name}{arg}...}", and optional args on acl condition
1706 JH/05 Permit multiple router/transport headers_add/remove lines.
1708 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1710 JH/07 Avoid using a waiting database for a single-message-only transport.
1711 Performance patch from Paul Fisher. Bugzilla 1262.
1713 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1716 JH/09 Add $headers_added variable, with content from use of ACL modifier
1717 add_header (but not yet added to the message). Bugzilla 199.
1719 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1720 Pulled from Bugzilla 817 by Wolfgang Breyha.
1722 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1724 (nb: this is the same fix as in Exim 4.80.1)
1726 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1729 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1732 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1734 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1735 advertise SMTP AUTH mechanism to us, instead of a generic
1736 protocol violation error. Also, make Exim more robust to bad
1737 data from the Dovecot auth socket.
1739 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1741 When a queue runner is handling a message, Exim first routes the
1742 recipient addresses, during which it prunes them based on the retry
1743 hints database. After that it attempts to deliver the message to
1744 any remaining recipients. It then updates the hints database using
1747 So if a recipient address works intermittently, it can get repeatedly
1748 deferred at routing time. The retry hints record remains fresh so the
1749 address never reaches the final cutoff time.
1751 This is a fairly common occurrence when a user is bumping up against
1752 their storage quota. Exim had some logic in its local delivery code
1753 to deal with this. However it did not apply to per-recipient defers
1754 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1756 This change adds a proper retry rule check during routing so that the
1757 final cutoff time is checked against the message's age. We only do
1758 this check if there is an address retry record and there is not a
1759 domain retry record; this implies that previous attempts to handle
1760 the address had the retry_use_local_parts option turned on. We use
1761 this as an approximation for the destination being like a local
1762 delivery, as in LMTP.
1764 I suspect this new check makes the old local delivery cutoff check
1765 redundant, but I have not verified this so I left the code in place.
1767 TF/02 Correct gecos expansion when From: is a prefix of the username.
1769 Test 0254 submits a message to Exim with the header
1773 When I ran the test suite under the user fanf2, Exim expanded
1774 the header to contain my full name, whereas it should have added
1775 a Resent-Sender: header. It erroneously treats any prefix of the
1776 username as equal to the username.
1778 This change corrects that bug.
1780 GF/01 DCC debug and logging tidyup
1781 Error conditions log to paniclog rather than rejectlog.
1782 Debug lines prefixed by "DCC: " to remove any ambiguity.
1784 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1786 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1787 Bug spotted by Jeremy Harris; was flawed since initial commit.
1788 Would have resulted in OCSP responses post-SNI triggering an Exim
1789 NULL dereference and crash.
1791 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1793 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1794 Bug detection, analysis and fix by Samuel Thibault.
1795 Bugzilla 1331, Debian bug #698092.
1797 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1799 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1800 Server implementation by Todd Lyons, client by JH.
1801 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1802 config variable "prdr_enable" controls whether the server
1803 advertises the facility. If the client requests PRDR a new
1804 acl_data_smtp_prdr ACL is called once for each recipient, after
1805 the body content is received and before the acl_smtp_data ACL.
1806 The client is controlled by both of: a hosts_try_prdr option
1807 on the smtp transport, and the server advertisement.
1808 Default client logging of deliveries and rejections involving
1809 PRDR are flagged with the string "PRDR".
1811 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1812 fclose(). Diagnosis by Todd Lyons.
1814 PP/17 Update configure.default to handle IPv6 localhost better.
1815 Patch by Alain Williams (plus minor tweaks).
1818 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1819 This is now consistent with GnuTLS, and is now documented: the
1820 previous undocumented portable approach to treating the option as
1821 unset was to force an expansion failure. That still works, and
1822 an empty string is now equivalent.
1824 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1825 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1826 not performing validation itself.
1828 PP/20 Added force_command boolean option to pipe transport.
1829 Patch from Nick Koston, of cPanel Inc.
1831 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1834 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1836 PP/21 Fix eximon continuous updating with timestamped log-files.
1837 Broken in a format-string cleanup in 4.80, missed when I repaired the
1838 other false fix of the same issue.
1839 Report and fix from Heiko Schlichting.
1842 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1843 Report from Prashanth Katuri.
1845 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1846 It's SecureTransport, so affects any MacOS clients which use the
1847 system-integrated TLS libraries, including email clients.
1849 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1850 using a MIME ACL for non-SMTP local injection.
1851 Report and assistance in diagnosis by Warren Baker.
1853 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1855 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1857 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1858 last id that failed so it may be referenced in subsequent ACL's.
1860 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1863 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1864 ldap library initialization, allowing self-signed CA's to be
1865 used. Also properly sets require_cert option later in code by
1866 using NULL (global ldap config) instead of ldap handle (per
1867 session). Bug diagnosis and testing by alxgomz.
1869 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1870 the src/util/ subdirectory.
1872 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1873 renamed to Transport Post Delivery Action by Jeremy Harris, as
1876 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1877 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1878 redis_servers = needs to be configured which will be used by the redis
1879 lookup. Patch from Warren Baker, of The Packet Hub.
1881 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1883 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1884 hostname or reverse DNS when processing a host list. Used suggestions
1885 from multiple comments on this bug.
1887 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1889 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1890 Missed a few lines, added it to make the runtest require no keyboard
1893 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1894 contains upper case chars. Make router use caseful_local_part.
1896 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1897 support when GnuTLS has been built with p11-kit.
1903 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1905 This, or similar/improved, will also be change PP/11 of 4.82.
1911 PP/01 Handle short writes when writing local log-files.
1912 In practice, only affects FreeBSD (8 onwards).
1913 Bugzilla 1053, with thanks to Dmitry Isaikin.
1915 NM/01 Bugzilla 949 - Documentation tweak
1917 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1920 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1922 PP/02 Implemented gsasl authenticator.
1924 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1926 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1927 `pkg-config foo` for cflags/libs.
1929 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1930 with rest of GSASL and with heimdal_gssapi.
1932 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1933 `pkg-config foo` for cflags/libs for the TLS implementation.
1935 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1936 properties get this fed in as external SSF. A number of robustness
1937 and debugging improvements to the cyrus_sasl authenticator.
1939 PP/08 cyrus_sasl server now expands the server_realm option.
1941 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1942 Patch by Jeremy Harris.
1944 PP/10 Added dbmjz lookup type.
1946 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1948 PP/12 MAIL args handles TAB as well as SP, for better interop with
1949 non-compliant senders.
1950 Analysis and variant patch by Todd Lyons.
1952 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1953 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1954 Patch from Dirk Mueller <dmueller@suse.com>
1956 PP/13 tls_peerdn now print-escaped for spool files.
1957 Observed some $tls_peerdn in wild which contained \n, which resulted
1958 in spool file corruption.
1960 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1961 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1962 or write after TLS renegotiation, which otherwise led to messages
1965 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1966 as a tracking header (ie: a signed header comes before the signature).
1967 Patch from Wolfgang Breyha.
1969 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1970 comma-sep list; embedded commas doubled.
1972 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1974 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1976 Report and patch from Dmitry Banschikov.
1978 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1979 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1980 failures. We appear to now support TLS1.1+ with Exim.
1982 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1983 lets Exim select keys and certificates based upon TLS SNI from client.
1984 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1985 before an outbound SMTP session. New log_selector, +tls_sni.
1987 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1988 NULL dereference. Report and patch from Alun Jones.
1990 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1991 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1992 Not seeing resolver debug output on NetBSD, but suspect this is a
1993 resolver implementation change.
1995 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1996 Left warnings. Added "eximon gdb" invocation mode.
1998 PP/21 Defaulting "accept_8bitmime" to true, not false.
2000 PP/22 Added -bw for inetd wait mode support.
2002 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
2003 locate the relevant includes and libraries. Made this the default.
2005 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
2006 Bugzilla 1246, report and most of solution from Tomasz Kusy.
2008 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
2009 This may cause build issues on older platforms.
2011 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
2012 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
2013 gnutls_require_mac & gnutls_require_protocols (no longer supported).
2014 Added SNI support via GnuTLS too.
2015 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
2017 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
2019 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
2020 Applied second patch from Janne, implementing suggestion to default
2021 multiple-strings-in-record handling to match SPF spec.
2023 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
2025 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
2026 read-only, out of scope).
2027 Patch from Wolfgang Breyha, report from Stuart Northfield.
2029 PP/29 Fix three issues highlighted by clang analyser static analysis.
2030 Only crash-plausible issue would require the Cambridge-specific
2031 iplookup router and a misconfiguration.
2032 Report from Marcin Mirosław.
2034 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
2036 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
2037 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
2038 As part of this, removing so much warning spew let me fix some minor
2039 real issues in debug logging.
2041 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
2042 assignment on my part. Fixed.
2044 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
2045 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
2046 Janne Snabb (who went above and beyond: thank you).
2048 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
2049 string otherwise requires a connection and a bunch more work and it's
2050 relatively easy to get wrong. Should also expose TLS library linkage
2053 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
2054 64-bit ${eval} (JH/03).
2056 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
2057 GNU libc to support some of the 64-bit stuff, should not lead to
2058 conflicts. Defined before os.h is pulled in, so if a given platform
2059 needs to override this, it can.
2061 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
2062 protection layer was required, which is not implemented.
2063 Bugzilla 1254, patch from Wolfgang Breyha.
2065 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
2066 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
2067 tls_dhparam take prime identifiers. Also unbreak combination of
2068 OpenSSL+DH_params+TLSSNI.
2070 PP/39 Disable SSLv2 by default in OpenSSL support.
2076 PP/01 Solaris build fix for Oracle's LDAP libraries.
2077 Bugzilla 1109, patch from Stephen Usher.
2079 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
2081 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
2084 TF/02 Fix a couple more cases where we did not log the error message
2085 when unlink() failed. See also change 4.74-TF/03.
2087 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
2088 lock up or crash if it happened to be inside a call to libc when it
2089 got a SIGUSR1 from exiwhat.
2091 The SIGUSR1 handler appends the current process status to the process
2092 log which is later printed by exiwhat. It used to use the general
2093 purpose logging code to do this, but several functions it calls are
2094 not safe for signals.
2096 The new output code in the SIGUSR1 handler is specific to the process
2097 log, and simple enough that it's easy to inspect for signal safety.
2098 Removing some special cases also simplifies the general logging code.
2099 Removing the spurious timestamps from the process log simplifies
2102 TF/04 Improved ratelimit ACL condition.
2104 The /noupdate option has been deprecated in favour of /readonly which
2105 has clearer semantics. The /leaky, /strict, and /readonly update modes
2106 are mutually exclusive. The update mode is no longer included in the
2107 database key; it just determines when the database is updated. (This
2108 means that when you upgrade Exim will forget old rate measurements.)
2110 Exim now checks that the per_* options are used with an update mode that
2111 makes sense for the current ACL. For example, when Exim is processing a
2112 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
2113 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
2114 must specify per_mail/readonly. If you omit the update mode it defaults to
2115 /leaky where that makes sense (as before) or /readonly where required.
2117 The /noupdate option is now undocumented but still supported for
2118 backwards compatibility. It is equivalent to /readonly except that in
2119 ACLs where /readonly is required you may specify /leaky/noupdate or
2120 /strict/noupdate which are treated the same as /readonly.
2122 A useful new feature is the /count= option. This is a generalization
2123 of the per_byte option, so that you can measure the throughput of other
2124 aggregate values. For example, the per_byte option is now equivalent
2125 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
2127 The per_rcpt option has been generalized using the /count= mechanism
2128 (though it's more complicated than the per_byte equivalence). When it is
2129 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
2130 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
2131 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
2132 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
2133 Note that using per_rcpt with a non-readonly update mode in more than
2134 one ACL will cause the recipients to be double-counted. (The per_mail
2135 and per_byte options don't have this problem.)
2137 The handling of very low rates has changed slightly. If the computed rate
2138 is less than the event's count (usually one) then this event is the first
2139 after a long gap. In this case the rate is set to the same as this event's
2140 count, so that the first message of a spam run is counted properly.
2142 The major new feature is a mechanism for counting the rate of unique
2143 events. The new per_addr option counts the number of different
2144 recipients that someone has sent messages to in the last time period. It
2145 behaves like per_rcpt if all the recipient addresses are different, but
2146 duplicate recipient addresses do not increase the measured rate. Like
2147 the /count= option this is a general mechanism, so the per_addr option
2148 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
2149 example, measure the rate that a client uses different sender addresses
2150 with the options per_mail/unique=$sender_address. There are further
2151 details in the main documentation.
2153 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
2155 TF/06 Removed a few PCRE remnants.
2157 TF/07 Automatically extract Exim's version number from tags in the git
2158 repository when doing development or release builds.
2160 PP/02 Raise smtp_cmd_buffer_size to 16kB.
2161 Bugzilla 879. Patch from Paul Fisher.
2163 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
2164 Heavily based on revision 40f9a89a from Simon Arlott's tree.
2167 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
2169 PP/05 Variable $av_failed, true if the AV scanner deferred.
2170 Bugzilla 1078. Patch from John Horne.
2172 PP/06 Stop make process more reliably on build failure.
2173 Bugzilla 1087. Patch from Heiko Schlittermann.
2175 PP/07 Make maildir_use_size_file an _expandable_ boolean.
2176 Bugzilla 1089. Patch from Heiko Schlittermann.
2178 PP/08 Handle ${run} returning more data than OS pipe buffer size.
2179 Bugzilla 1131. Patch from Holger Weiß.
2181 PP/09 Handle IPv6 addresses with SPF.
2182 Bugzilla 860. Patch from Wolfgang Breyha.
2184 PP/10 GnuTLS: support TLS 1.2 & 1.1.
2186 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
2189 PP/11 match_* no longer expand right-hand-side by default.
2190 New compile-time build option, EXPAND_LISTMATCH_RHS.
2191 New expansion conditions, "inlist", "inlisti".
2193 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
2195 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
2197 PP/14 fix log_write() format string regression from TF/03.
2198 Bugzilla 1152. Patch from Dmitry Isaikin.
2204 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
2206 PP/02 Harmonised TLS library version reporting; only show if debugging.
2207 Layout now matches that introduced for other libraries in 4.74 PP/03.
2209 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
2211 PP/04 New "dns_use_edns0" global option.
2213 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
2216 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
2217 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
2219 TK/01 Updated PolarSSL code to 0.14.2.
2220 Bugzilla 1097. Patch from Andreas Metzler.
2222 PP/07 Catch divide-by-zero in ${eval:...}.
2223 Fixes bugzilla 1102.
2225 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
2228 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
2229 format-string attack -- SECURITY: remote arbitrary code execution.
2231 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
2232 time unintentionally subject to list matching rules, letting the header
2233 cause arbitrary Exim lookups (of items which can occur in lists, *not*
2234 arbitrary string expansion). This allowed for information disclosure.
2236 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
2237 INT_MIN/-1 -- value coerced to INT_MAX.
2243 NM/01 Workaround for PCRE version dependency in version reporting
2246 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
2247 This fixes portability to compilers other than gcc, notably
2248 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
2250 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
2251 makefiles for portability to HP-UX and POSIX correctness.
2253 PP/01 Permit LOOKUP_foo enabling on the make command-line.
2254 Also via indented variable definition in the Makefile.
2255 (Debugging by Oliver Heesakkers).
2257 PP/02 Restore caching of spamd results with expanded spamd_address.
2258 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
2260 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
2261 Improves build reliability. Fix from: Frank Elsner
2263 NM/02 Fix wide character breakage in the rfc2047 coding
2264 Fixes bug 1064. Patch from Andrey N. Oktyabrski
2266 NM/03 Allow underscore in dnslist lookups
2267 Fixes bug 1026. Patch from Graeme Fowler
2269 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
2270 Code patches from Adam Ciarcinski of NetBSD.
2272 NM/04 Fixed exiqgrep to cope with mailq missing size issue
2275 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
2276 is logged, to avoid truncation. Patch from John Horne.
2278 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
2279 Patch from Jakob Hirsch.
2281 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
2282 SQL string expansion failure details.
2283 Patch from Andrey Oktyabrski.
2285 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
2286 Patch from Simon Arlott.
2288 PP/09 New lookups functionality failed to compile on old gcc which rejects
2289 extern declarations in function scope.
2290 Patch from Oliver Fleischmann
2292 PP/10 Use sig_atomic_t for flags set from signal handlers.
2293 Check getgroups() return and improve debugging.
2294 Fixed developed for diagnosis in bug 927 (which turned out to be
2297 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
2298 Patch from Mark Zealey.
2300 PP/12 Bugzilla 1056: Improved spamd server selection.
2301 Patch from Mark Zealey.
2303 PP/13 Bugzilla 1086: Deal with maildir quota file races.
2304 Based on patch from Heiko Schlittermann.
2306 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
2307 Patch from Uwe Doering, sign-off by Michael Haardt.
2309 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
2310 variable declaration deep within a block. Bug and patch from
2313 PP/15 lookups-Makefile IRIX compatibility coercion.
2315 PP/16 Make DISABLE_DKIM build knob functional.
2317 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
2318 Patch by Simon Arlott
2320 TF/03 Fix valgrind.h portability to C89 compilers that do not support
2321 variable argument macros. Our copy now differs from upstream.
2327 TF/01 Failure to get a lock on a hints database can have serious
2328 consequences so log it to the panic log.
2330 TF/02 Log LMTP confirmation messages in the same way as SMTP,
2331 controlled using the smtp_confirmation log selector.
2333 TF/03 Include the error message when we fail to unlink a spool file.
2335 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
2336 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
2337 for maintaining out-of-tree patches for some time.
2339 PP/01 Bugzilla 139: Documentation and portability issues.
2340 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
2341 Handle per-OS dynamic-module compilation flags.
2343 PP/02 Let /dev/null have normal permissions.
2344 The 4.73 fixes were a little too stringent and complained about the
2345 permissions on /dev/null. Exempt it from some checks.
2346 Reported by Andreas M. Kirchwitz.
2348 PP/03 Report version information for many libraries, including
2349 Exim version information for dynamically loaded libraries. Created
2350 version.h, now support a version extension string for distributors
2351 who patch heavily. Dynamic module ABI change.
2353 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
2354 privilege escalation vulnerability whereby the Exim run-time user
2355 can cause root to append content of the attacker's choosing to
2358 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
2361 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
2362 If dropping privileges for untrusted macros, we disabled normal logging
2363 on the basis that it would fail; for the Exim run-time user, this is not
2364 the case, and it resulted in successful deliveries going unlogged.
2365 Fixed. Reported by Andreas Metzler.
2371 PP/01 Date: & Message-Id: revert to normally being appended to a message,
2372 only prepend for the Resent-* case. Fixes regression introduced in
2373 Exim 4.70 by NM/22 for Bugzilla 607.
2375 PP/02 Include check_rfc2047_length in configure.default because we're seeing
2376 increasing numbers of administrators be bitten by this.
2378 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
2380 PP/03 Bugzilla 994: added openssl_options main configuration option.
2382 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
2384 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
2386 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
2388 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
2389 without a peer certificate, leading to a segfault because of an
2390 assumption that peers always have certificates. Be a little more
2391 paranoid. Problem reported by Martin Tscholak.
2393 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
2394 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
2395 NB: ClamAV planning to remove STREAM in "middle of 2010".
2396 CL also introduces -bmalware, various -d+acl logging additions and
2397 more caution in buffer sizes.
2399 PP/09 Implemented reverse_ip expansion operator.
2401 PP/10 Bugzilla 937: provide a "debug" ACL control.
2403 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
2405 PP/12 Bugzilla 973: Implement --version.
2407 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
2409 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
2411 PP/15 Bugzilla 816: support multiple condition rules on Routers.
2413 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
2414 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
2415 ignore trailing whitespace.
2417 JJ/02 prevent non-panic DKIM error from being sent to paniclog
2419 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
2422 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
2423 Notification from Dr Andrew Aitchison.
2425 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
2426 ExtendedDetectionInfo response format.
2427 Notification from John Horne.
2429 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
2432 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
2433 XSL and documented dependency on system catalogs, with examples of how
2436 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
2439 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
2440 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
2441 configuration file which is writeable by the Exim user or group.
2443 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
2444 of configuration files to cover files specified with the -C option if
2445 they are going to be used with root privileges, not just the default
2448 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
2449 option (effectively making it always true).
2451 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
2452 files to be used while preserving root privileges.
2454 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
2455 that rogue child processes cannot use them.
2457 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
2458 run-time user, instead of root.
2460 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
2461 Exim run-time user without dropping privileges.
2463 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
2464 result string, instead of calling string_vformat() twice with the same
2467 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
2468 for other users. Others should always drop root privileges if they use
2469 -C on the command line, even for a whitelisted configure file.
2471 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
2473 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2479 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2480 $data_path, and $header_path variables; fixed documentation bugs and
2483 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2484 exipick to access non-standard spools, including the "frozen" queue
2487 NM/01 Bugzilla 965: Support mysql stored procedures.
2488 Patch from Alain Williams
2490 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2492 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2493 Patch from Andreas Metzler
2495 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2496 Patch from Kirill Miazine
2498 NM/05 Bugzilla 671: Added umask to procmail example.
2500 JJ/03 installed exipick 20100323.0, fixing doc bug
2502 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2503 directory. Notification and patch from Dan Rosenberg.
2505 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2507 TK/02 Improve log output when DKIM signing operation fails.
2509 MH/01 Treat the transport option dkim_domain as a colon separated
2510 list, not as a single string, and sign the message with each element,
2511 omitting multiple occurences of the same signer.
2513 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2514 Bugzilla 985, 986. Patch by Simon Arlott
2516 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2517 Patch by Simon Arlott
2519 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2520 MBX locking. Notification from Dan Rosenberg.
2526 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2528 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2530 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2532 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2534 NM/04 Bugzilla 915: Flag broken perl installation during build.
2540 TK/01 Added patch by Johannes Berg that expands the main option
2541 "spamd_address" if it starts with a dollar sign.
2543 TK/02 Write list of recipients to X-Envelope-Sender header when building
2544 the mbox-format spool file for content scanning (suggested by Jakob
2547 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2548 (http://www.dcc-servers.net/) support via dccifd. Activated by
2549 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2551 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2552 by Mark Daniel Reidel <mr@df.eu>.
2554 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2555 When building exim an external PCRE library is now needed -
2556 PCRE is a system library on the majority of modern systems.
2557 See entry on PCRE_LIBS in EDITME file.
2559 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2560 conversation. Added nologin parameter to request.
2561 Patch contributed by Kirill Miazine.
2563 TF/01 Do not log submission mode rewrites if they do not change the address.
2565 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2567 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2568 log files in place. Contributed by Roberto Lima.
2570 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2572 TF/03 Bugzilla 615: When checking the local_parts router precondition
2573 after a local_part_suffix or local_part_prefix option, Exim now
2574 does not use the address's named list lookup cache, since this
2575 contains cached lookups for the whole local part.
2577 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2578 Robert Millan. Documentation is in experimental-spec.txt.
2580 TF/04 Bugzilla 668: Fix parallel build (make -j).
2582 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2584 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2585 Patch provided by Jan Srzednicki.
2587 TF/05 Leading white space used to be stripped from $spam_report which
2588 wrecked the formatting. Now it is preserved.
2590 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2591 that they are available at delivery time.
2593 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2595 TF/08 TLS error reporting now respects the incoming_interface and
2596 incoming_port log selectors.
2598 TF/09 Produce a more useful error message if an SMTP transport's hosts
2599 setting expands to an empty string.
2601 NM/06 Bugzilla 744: EXPN did not work under TLS.
2602 Patch provided by Phil Pennock.
2604 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2605 Patch provided by Richard Godbee.
2607 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2608 acl_smtp_notquit, added index entry.
2610 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2611 Patch provided by Eugene Bujak.
2613 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2614 accept(). Patch provided by Maxim Dounin.
2616 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2617 Patch provided by Phil Pennock.
2619 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2621 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2622 Patch provided by Brad "anomie" Jorsch.
2624 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2625 Patch provided by Dean Brooks.
2627 TK/05 Add native DKIM support (does not depend on external libraries).
2629 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2630 Patch provided by Graeme Fowler.
2632 NM/16 Bugzilla 851: Documentation example syntax fix.
2634 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2636 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2639 NM/19 Bugzilla 745: TLS version reporting.
2640 Patch provided by Phil Pennock.
2642 NM/20 Bugzilla 167: bool: condition support.
2643 Patch provided by Phil Pennock.
2645 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2646 clients. Patch provided by Phil Pennock.
2648 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2649 Patch provided by Brad "anomie" Jorsch.
2651 NM/23 Bugzilla 687: Fix misparses in eximstats.
2652 Patch provided by Heiko Schlittermann.
2654 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2655 Patch provided by Heiko Schlittermann.
2657 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2658 plus update to original patch.
2660 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2662 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2663 Patch provided by David Brownlee.
2665 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2667 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2669 NM/30 Bugzilla 888: TLS documentation bugfixes.
2671 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2673 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2674 Unlike the original bugzilla I have changed all shell scripts in src tree.
2676 NM/33 Bugzilla 898: Transport filter timeout fix.
2677 Patch by Todd Rinaldo.
2679 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2680 Patch by Serge Demonchaux.
2682 NM/35 Bugzilla 39: Base64 decode bug fixes.
2683 Patch by Jakob Hirsch.
2685 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2687 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2689 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2691 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2697 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2698 ALT-N's libdkim that I have put here:
2699 http://duncanthrax.net/exim-experimental/
2701 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2702 were called 'true' and it seems that C99 defines that as a reserved
2703 keyword to be used with 'bool' variable types. That means you could
2704 not include C99-style headers which use bools without triggering
2705 build errors in sieve.c.
2707 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2708 as mailq or other aliases. Changed the --help handling significantly
2709 to do whats expected. exim_usage() emits usage/help information.
2711 SC/01 Added the -bylocaldomain option to eximstats.
2713 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2715 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2717 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2723 PH/01 Another patch from the Sieve maintainer.
2725 PH/02 When an IPv6 address is converted to a string for single-key lookup
2726 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2727 dots are used instead of colons so that keys in lsearch files need not
2728 contain colons. This was done some time before quoting was made available
2729 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2730 (notated using the quote facility) so as to distinguish them from IPv4
2731 keys. This meant that lookups for IP addresses in host lists did not work
2732 for iplsearch lookups.
2734 This has been fixed by arranging for IPv6 addresses to be expressed with
2735 colons if the lookup type is iplsearch. This is not incompatible, because
2736 previously such lookups could never work.
2738 The situation is now rather anomalous, since one *can* have colons in
2739 ordinary lsearch keys. However, making the change in all cases is
2740 incompatible and would probably break a number of configurations.
2742 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2745 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2746 conversion specification without a maximum field width, thereby enabling
2747 a rogue spamd server to cause a buffer overflow. While nobody in their
2748 right mind would setup Exim to query an untrusted spamd server, an
2749 attacker that gains access to a server running spamd could potentially
2750 exploit this vulnerability to run arbitrary code as the Exim user.
2752 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2753 $primary_hostname instead of what libspf2 thinks the hosts name is.
2755 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2756 a directory entry by the name of the lookup key. Previously, if a
2757 symlink pointed to a non-existing file or a file in a directory that
2758 Exim lacked permissions to read, a lookup for a key matching that
2759 symlink would fail. Now it is enough that a matching directory entry
2760 exists, symlink or not. (Bugzilla 503.)
2762 PH/03 The body_linecount and body_zerocount variables are now exported in the
2765 PH/04 Added the $dnslist_matched variable.
2767 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2768 This means they are set thereafter only if the connection becomes
2771 PH/06 Added the client_condition to authenticators so that some can be skipped
2772 by clients under certain conditions.
2774 PH/07 The error message for a badly-placed control=no_multiline_responses left
2775 "_responses" off the end of the name.
2777 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2779 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2780 (without spaces) instead of just copying the configuration text.
2782 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2784 PH/11 Added $max_received_linelength.
2786 PH/12 Added +ignore_defer and +include_defer to host lists.
2788 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2789 way in which PCRE > 7.0 is built.
2791 PH/14 Implemented queue_only_load_latch.
2793 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2794 MAIL command. The effect was to mangle the value on 64-bit systems.
2796 PH/16 Another patch from the Sieve maintainer.
2798 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2800 PH/18 If a system quota error occurred while trying to create the file for
2801 a maildir delivery, the message "Mailbox is full" was not appended to the
2802 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2803 only to a quota excession during the actual writing of the file.
2805 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2806 characters?) which causes problems in log lines. The DN values are now
2807 passed through string_printing() before being added to log lines.
2809 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2810 and InterBase are left for another time.)
2812 PH/21 Added message_body_newlines option.
2814 PH/22 Guard against possible overflow in moan_check_errorcopy().
2816 PH/23 POSIX allows open() to be a macro; guard against that.
2818 PH/24 If the recipient of an error message contained an @ in the local part
2819 (suitably quoted, of course), incorrect values were put in $domain and
2820 $local_part during the evaluation of errors_copy.
2826 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2827 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2830 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2831 issue a MAIL command.
2833 PH/02 In an ACL statement such as
2835 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2837 if a client was not listed at all, or was listed with a value other than
2838 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2839 the condition was not true (as it should be), so access was not denied.
2840 The bug was that the ! inversion was incorrectly passed on to the second
2841 item. This has been fixed.
2843 PH/03 Added additional dnslists conditions == and =& which are different from
2844 = and & when the dns lookup returns more than one IP address.
2846 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2847 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2849 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2850 FSYNC, which compiles an option called disable_fsync that allows for
2851 bypassing fsync(). The documentation is heavily laced with warnings.
2853 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2855 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2856 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2857 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2858 including adding "make clean"; (3) Added -fPIC when compiling the test
2859 dynamically loaded module, to get rid of a warning.
2861 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2862 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2863 = 0s. The bug is otherwise harmless.
2865 PH/07 There was a bug in the dovecot authenticator such that the value of
2866 $auth1 could be overwritten, and so not correctly preserved, after a
2867 successful authentication. This usually meant that the value preserved by
2868 the server_setid option was incorrect.
2870 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2872 PH/09 Installed PCRE release 7.0.
2874 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2875 run for batched SMTP input. It is now run at the start of every message
2876 in the batch. While fixing this I discovered that the process information
2877 (output by running exiwhat) was not always getting set for -bs and -bS
2878 input. This is fixed, and it now also says "batched" for BSMTP.
2880 PH/11 Added control=no_pipelining.
2882 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2883 patch, slightly modified), and move the expansion of helo_data till after
2884 the connection is made in the smtp transport (so it can use these
2887 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2889 PH/14 Added log_selector = +pid.
2891 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2893 PH/16 Add ${if forany and ${if forall.
2895 PH/17 Added dsn_from option to vary the From: line in DSNs.
2897 PH/18 Flush SMTP output before performing a callout, unless control =
2898 no_callout_flush is set.
2900 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2901 was true (the default) a successful delivery failed to delete the retry
2902 item, thus causing premature timeout of the address. The bug is now
2905 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2907 PH/21 Long custom messages for fakedefer and fakereject are now split up
2908 into multiline responses in the same way that messages for "deny" and
2909 other ACL rejections are.
2911 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2912 with slight modification.
2914 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2915 draft, changing the syntax and factoring some duplicate code".
2917 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2918 for deliveries of the second and subsequent messages over the same SMTP
2921 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2922 ${reduce, with only minor "tidies".
2924 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2926 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2927 expansion side effects.
2929 PH/27 When a message times out after an over-quota error from an Exim-imposed
2930 quota, the bounce message says "mailbox is full". This message was not
2931 being given when it was a system quota that was exceeded. It now should
2934 MH/03 Made $recipients available in local_scan(). local_scan() already has
2935 better access to the recipient list through recipients_list[], but
2936 $recipients can be useful in postmaster-provided expansion strings.
2938 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2939 in the case of a MAIL command with additional options following the
2940 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2941 were accidentally chopped off.
2943 PH/29 SMTP synchronization checks are implemented when a command is read -
2944 there is a check that no more input is waiting when there shouldn't be
2945 any. However, for some commands, a delay in an ACL can mean that it is
2946 some time before the response is written. In this time, more input might
2947 arrive, invalidly. So now there are extra checks after an ACL has run for
2948 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2949 pipelining has not been advertised.
2951 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2953 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2954 returned in the SMTP response when smtp_return_error_details was false.
2955 This has been fixed.
2957 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2958 instead of the C I/O that was originally supplied, because problems were
2959 reported on Solaris.
2961 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2962 Exim which did not show up earlier: it was assuming that a call to
2963 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2964 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2965 was a macro that became an assignment, so it seemed to work. This has
2966 changed to a proper function call with a void return, hence the compile
2967 error. Exim's code has been fixed.
2969 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2972 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2974 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2976 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2977 because it turns out to be too big or there is a timeout) there is no
2978 "Completed" line in the log. When some messages of this type were
2979 selected by exigrep, they were listed as "not completed". Others were
2980 picked up by some special patterns. I have improved the selection
2981 criteria to be more general.
2983 PH/38 The host_find_failed option in the manualroute router can now be set
2984 to "ignore", to completely ignore a host whose IP address cannot be
2985 found. If all hosts are ignored, the behaviour is controlled by the new
2986 host_all_ignored option.
2988 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2989 homing or because of multiple MX records with /mx) generated more than
2990 one IP address, and the following item turned out to be the local host,
2991 all the secondary addresses of the first item were incorrectly removed
2992 from the list, along with the local host and any following hosts (which
2993 is what is supposed to happen).
2995 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2996 whoever called Exim into the -H file. In the case of the daemon it was
2997 behaving confusingly. When first started, it used values for whoever
2998 started the daemon, but after a SIGHUP it used the Exim user (because it
2999 calls itself on a restart). I have changed the code so that it now always
3002 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
3003 message are rejected with the same error (e.g. no authentication or bad
3004 sender address), and a DATA command is nevertheless sent (as can happen
3005 with PIPELINING or a stupid MUA), the error message that was given to the
3006 RCPT commands is included in the rejection of the DATA command. This is
3007 intended to be helpful for MUAs that show only the final error to their
3010 PH/42 Another patch from the Sieve maintainer.
3012 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
3013 Eximstats - Fixed some broken HTML links and added missing column headers
3015 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
3016 columns for Rejects, Temp Rejects, Ham, and Spam rows.
3018 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
3020 PH/43 Yet another patch from the Sieve maintainer.
3022 PH/44 I found a way to check for a TCP/IP connection going away before sending
3023 the response to the final '.' that terminates a message, but only in the
3024 case where the client has not sent further data following the '.'
3025 (unfortunately, this is allowed). However, in many cases there won't be
3026 any further data because there won't be any more messages to send. A call
3027 to select() can be used: if it shows that the input is "ready", there is
3028 either input waiting, or the socket has been closed. An attempt to read
3029 the next input character can distinguish the two cases. Previously, Exim
3030 would have sent an OK response which the client would never have see.
3031 This could lead to message repetition. This fix should cure that, at
3032 least in a lot of common cases.
3034 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
3035 advertised in response to EHLO.
3041 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
3042 fixed by 4.65/MH/01 (is this a record?) are fixed:
3044 (i) An empty string was always treated as zero by the numeric comparison
3045 operators. This behaviour has been restored.
3047 (ii) It is documented that the numeric comparison operators always treat
3048 their arguments as decimal numbers. This was broken in that numbers
3049 starting with 0 were being interpreted as octal.
3051 While fixing these problems I realized that there was another issue that
3052 hadn't been noticed. Values of message_size_limit (both the global option
3053 and the transport option) were treated as octal if they started with 0.
3054 The documentation was vague. These values are now always treated as
3055 decimal, and I will make that clear in the documentation.
3061 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
3062 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
3065 MH/01 Don't check that the operands of numeric comparison operators are
3066 integers when their expansion is in "skipping" mode (fixes bug
3067 introduced by 4.64-PH/07).
3069 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
3070 child addresses, Exim now panics and dies. Previously, because the count
3071 is held in a short int, deliveries were likely to be lost. As such a
3072 large number of recipients for a single message is ridiculous
3073 (performance will be very, very poor), I have chosen to impose a limit
3074 rather than extend the field.
3080 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
3081 leftover -K file (the existence of which was triggered by #402).
3082 While we were at it, introduced process PID as part of the -K
3083 filename. This should rule out race conditions when creating
3086 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
3087 processing considerably. Previous code took too long for large mails,
3088 triggering a timeout which in turn triggers #401.
3090 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
3091 in the DK code in transports.c. sendfile() is not really portable,
3092 hence the _LINUX specificness.
3094 TF/01 In the add_headers option to the mail command in an Exim filter,
3095 there was a bug that Exim would claim a syntax error in any
3096 header after the first one which had an odd number of characters
3099 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
3100 callout verification, Exim cached a "reject" for the entire domain. This
3101 is correct for most verifications, but it is not correct for a recipient
3102 verification with use_sender or use_postmaster set, because in that case
3103 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
3104 case of MAIL FROM:<> rejection from other early rejections (e.g.
3105 rejection of HELO). When verifying a recipient using a non-null MAIL
3106 address, the cache is ignored if it shows MAIL FROM:<> rejection.
3107 Whatever the result of the callout, the value of the domain cache is
3108 left unchanged (for any other kind of callout, getting as far as trying
3109 RCPT means that the domain itself is ok).
3111 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
3114 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
3115 manifest itself as EPIPE rather than ECONNECT. When tidying away a
3116 session, the daemon ignores ECONNECT errors and logs others; it now
3117 ignores EPIPE as well.
3119 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
3120 (quoted-printable decoding).
3122 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
3123 later the small subsequent patch to fix an introduced bug.
3125 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
3127 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
3129 PH/08 An error is now given if message_size_limit is specified negative.
3131 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
3132 to be given (somewhat) arbitrary names.
3134 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
3137 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
3138 miscellaneous code fixes
3140 PH/10 Added the log_reject_target ACL modifier to specify where to log
3143 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
3144 hostname. This is wrong, because it relates to the incoming message (and
3145 probably the interface on which it is arriving) and not to the outgoing
3146 callout (which could be using a different interface). This has been
3147 changed to use the value of the helo_data option from the smtp transport
3148 instead - this is what is used when a message is actually being sent. If
3149 there is no remote transport (possible with a router that sets up host
3150 addresses), $smtp_active_hostname is used.
3152 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
3153 tweaks were necessary in order to get it to work (see also 21 below):
3154 (a) The code assumed that strncpy() returns a negative number on buffer
3155 overflow, which isn't the case. Replaced with Exim's string_format()
3157 (b) There were several signed/unsigned issues. I just did the minimum
3158 hacking in of casts. There is scope for a larger refactoring.
3159 (c) The code used strcasecmp() which is not a standard C function.
3160 Replaced with Exim's strcmpic() function.
3161 (d) The code set only $1; it now sets $auth1 as well.
3162 (e) A simple test gave the error "authentication client didn't specify
3163 service in request". It would seem that Dovecot has changed its
3164 interface. Fortunately there's a specification; I followed it and
3165 changed what the client sends and it appears to be working now.
3167 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
3170 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
3171 address A is aliased to B and C, where B exists and C does not. Without
3172 -v the output is "A verified" because verification stops after a
3173 successful redirection if more than one address is generated. However,
3174 with -v the child addresses are also verified. Exim was outputting "A
3175 failed to verify" and then showing the successful verification for C,
3176 with its parentage. It now outputs "B failed to verify", showing B's
3177 parentage before showing the successful verification of C.
3179 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
3180 look up a TXT record in a specific list after matching in a combined
3183 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
3184 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
3185 they consult the DNS. I had assumed they would set it the way they
3186 wanted; and indeed my experiments on Linux seem to show that in some
3187 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
3188 To be on the safe side, however, I have now made the interface to
3189 host_find_byname() similar to host_find_bydns(), with an argument
3190 containing the DNS resolver options. The host_find_byname() function now
3191 sets these options at its start, just as host_find_bydns() does. The smtp
3192 transport options dns_qualify_single and dns_search_parents are passed to
3193 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
3194 of host_find_byname() use the default settings of RES_DEFNAMES
3195 (qualify_single) but not RES_DNSRCH (search_parents).
3197 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
3198 spool_read_header() do less string testing, by means of a preliminary
3199 switch on the second character of optional "-foo" lines. (This is
3200 overdue, caused by the large number of possibilities that now exist.
3201 Originally there were few.) While I was there, I also converted the
3202 str(n)cmp tests so they don't re-test the leading "-" and the first
3203 character, in the hope this might squeeze out yet more improvement.
3205 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
3206 flag allowing group syntax was set by the header_syntax check but not
3207 turned off, possible causing trouble later; (2) The flag was not being
3208 set at all for the header_verify test, causing "group"-style headers to
3209 be rejected. I have now set it in this case, and also caused header_
3210 verify to ignore an empty address taken from a group. While doing this, I
3211 came across some other cases where the code for allowing group syntax
3212 while scanning a header line wasn't quite right (mostly, not resetting
3213 the flag correctly in the right place). These bugs could have caused
3214 trouble for malformed header lines. I hope it is now all correct.
3216 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
3217 with the "reply" argument non-NULL. The code, however (which originally
3218 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
3219 but it didn't always do it. This confused somebody who was copying the
3220 code for some other use. I have removed all the tests.
3222 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
3223 feature that was used to support insecure browsers during the U.S. crypto
3224 embargo. It requires special client support, and Exim is probably the
3225 only MTA that supported it -- and would never use it because real RSA is
3226 always available. This code has been removed, because it had the bad
3227 effect of slowing Exim down by computing (never used) parameters for the
3228 RSA_EXPORT functionality.
3230 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
3231 authenticator to fail if there's a tab character in the incoming data
3232 (there should never be unless someone is messing about, as it's supposed
3233 to be base64-encoded). Also added, on Timo's advice, the "secured" option
3234 if the connection is using TLS or if the remote IP is the same as the
3235 local IP, and the "valid-client-cert option" if a client certificate has
3238 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
3239 authenticators. This can be used for authorization after authentication
3240 succeeds. (In the case of plaintext, it servers for both authentication
3243 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
3244 if any retry times were supplied.
3246 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
3247 connection, where there is no client IP address to check. In this
3248 situation, the verify now always succeeds.
3250 PH/25 Applied John Jetmore's -Mset patch.
3252 PH/26 Added -bem to be like -Mset, but loading a message from a file.
3254 PH/27 In a string expansion for a processed (not raw) header when multiple
3255 headers of the same name were present, leading whitespace was being
3256 removed from all of them, but trailing whitespace was being removed only
3257 from the last one. Now trailing whitespace is removed from each header
3258 before concatenation. Completely empty headers in a concatenation (as
3259 before) are ignored.
3261 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
3262 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
3264 PH/29 [Removed. This was a change that I later backed out, and forgot to
3265 correct the ChangeLog entry (that I had efficiently created) before
3266 committing the later change.]
3268 PH/30 Exim was sometimes attempting to deliver messages that had suffered
3269 address errors (4xx response to RCPT) over the same connection as other
3270 messages routed to the same hosts. Such deliveries are always "forced",
3271 so retry times are not inspected. This resulted in far too many retries
3272 for the affected addresses. The effect occurred only when there were more
3273 hosts than the hosts_max_try setting in the smtp transport when it had
3274 the 4xx errors. Those hosts that it had tried were not added to the list
3275 of hosts for which the message was waiting, so if all were tried, there
3276 was no problem. Two fixes have been applied:
3278 (i) If there are any address or message errors in an SMTP delivery, none
3279 of the hosts (tried or untried) are now added to the list of hosts
3280 for which the message is waiting, so the message should not be a
3281 candidate for sending over the same connection that was used for a
3282 successful delivery of some other message. This seems entirely
3283 reasonable: after all the message is NOT "waiting for some host".
3284 This is so "obvious" that I'm not sure why it wasn't done
3285 previously. Hope I haven't missed anything, but it can't do any
3286 harm, as the worst effect is to miss an optimization.
3288 (ii) If, despite (i), such a delivery is accidentally attempted, the
3289 routing retry time is respected, so at least it doesn't keep
3290 hammering the server.
3292 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
3293 in ${readsocket because some servers need this prod.
3295 PH/32 Added some extra debug output when updating a wait-xxx database.
3297 PH/33 The hint "could be header name not terminated by colon", which has been
3298 given for certain expansion errors for a long time, was not being given
3299 for the ${if def:h_colon_omitted{... case.
3301 PH/34 The spec says: "With one important exception, whenever a domain list is
3302 being scanned, $domain contains the subject domain." There was at least
3303 one case where this was not true.
3305 PH/35 The error "getsockname() failed: connection reset by peer" was being
3306 written to the panic log as well as the main log, but it isn't really
3307 panic-worthy as it just means the connection died rather early on. I have
3308 removed the panic log writing for the ECONNRESET error when getsockname()
3311 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
3312 runs only) independently of the message's sender address. This meant
3313 that, if the 4xx error was in fact related to the sender, a different
3314 message to the same recipient with a different sender could confuse
3315 things. In particular, this can happen when sending to a greylisting
3316 server, but other circumstances could also provoke similar problems.
3317 I have changed the default so that the retry time for these errors is now
3318 based a combination of the sender and recipient addresses. This change
3319 can be overridden by setting address_retry_include_sender=false in the
3322 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
3323 remote server are returned as part of bounce messages. This was not
3324 happening for LMTP over a pipe (the lmtp transport), but now it is the
3325 same for both kinds of LMTP.
3327 PH/38 Despite being documented as not happening, Exim was rewriting addresses
3328 in header lines that were in fact CNAMEs. This is no longer the case.
3330 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
3331 and queue runs started by the daemon processed all messages. This has
3332 been fixed so that -R and -S can now usefully be given with -q<time>.
3334 PH/40 Import PCRE release 6.7 (fixes some bugs).
3336 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
3338 PH/42 Give an error if -q is specified more than once.
3340 PH/43 Renamed the variables $interface_address and $interface_port as
3341 $received_ip_address and $received_port, to make it clear that these
3342 values apply to message reception, and not to the outgoing interface when
3343 a message is delivered. (The old names remain recognized, of course.)
3345 PH/44 There was no timeout on the connect() call when using a Unix domain
3346 socket in the ${readsocket expansion. There now is.
3348 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
3349 be meaningful with "accept".
3351 SC/01 Eximstats V1.43
3352 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
3354 SC/02 Eximstats V1.44
3355 Use a glob alias rather than an array ref in the generated
3356 parser. This improves both readability and performance.
3358 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
3359 Collect SpamAssassin and rejection statistics.
3360 Don't display local sender or destination tables unless
3361 there is data to show.
3362 Added average volumes into the top table text output.
3364 SC/04 Eximstats V1.46
3365 Collect data on the number of addresses (recipients)
3366 as well as the number of messages.
3368 SC/05 Eximstats V1.47
3369 Added 'Message too big' to the list of mail rejection
3370 reasons (thanks to Marco Gaiarin).
3372 SC/06 Eximstats V1.48
3373 Mainlog lines which have GMT offsets and are too short to
3374 have a flag are now skipped.
3376 SC/07 Eximstats V1.49 (Alain Williams)
3377 Added the -emptyok flag.
3379 SC/08 Eximstats V1.50
3380 Fixes for obtaining the IP address from reject messages.
3382 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
3383 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
3384 whitespace changes from 4.64-PH/27
3386 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
3389 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
3390 are found, allow negative numbers in numeric criteria)
3392 JJ/06 exipick.20061117.2, added new $message_body_missing variable
3394 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
3395 to match changes made in 4.64-PH/43
3397 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
3399 PH/47 Put in an explicit test for a DNS lookup of an address record where the
3400 "domain" is actually an IP address, and force a failure. This locks out
3401 those revolvers/nameservers that support "A-for-A" lookups, in
3402 contravention of the specifications.
3404 PH/48 When a host name was looked up from an IP address, and the subsequent
3405 forward lookup of the name timed out, the host name was left in
3406 $sender_host_name, contrary to the specification.
3408 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
3409 restricted to single-key lookups, Exim was not diagnosing an error if
3410 * or *@ was used with a query-style lookup.
3412 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
3414 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
3415 long ago, but noone interested enough thought of it. Let's just say that
3416 the "1.1" means that there are some new functions that weren't there at
3417 some point in the past.
3419 PH/51 Error processing for expansion failure of helo_data from an smtp
3420 transport during callout processing was broken.
3422 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
3423 tested/used via the -bh/-bhc/-bs options.
3425 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
3426 bug, fixed in subsequent PCRE releases).
3428 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
3429 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
3431 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
3437 SC/01 Use a glob alias rather than an array ref in eximstats generated
3438 parser. This improves both readability and performance.
3440 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
3441 Don't display local sender or destination tables in eximstats unless
3442 there is data to show.
3443 Added average volumes into the eximstats top table text output.
3445 SC/03 Collect data on the number of addresses (recipients) as well
3446 as the number of messages in eximstats.
3448 TF/01 Correct an error in the documentation for the redirect router. Exim
3449 does (usually) call initgroups() when daemonizing.
3451 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
3452 with consistent privilege compared to when running as a daemon.
3454 TF/03 Note in the spec that $authenticated_id is not set for local
3455 submissions from trusted users.
3457 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
3458 Thanks to Dean Brooks <dean@iglou.com> for the patch.
3460 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
3461 by adding some example configuration directives to the default
3462 configuration file. A little bit of work is required to uncomment the
3463 directives and define how usernames and passwords are checked, but
3464 there is now a framework to start from.
3466 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
3467 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
3468 without this. I don't know how relevant this is to other LDAP libraries.
3470 PH/02 Add the verb name to the "unknown ACL verb" error.
3472 PH/03 Magnus Holmgren's patch for filter_prepend_home.
3474 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
3476 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
3477 directory not expanded when it should be if an expanded home directory
3478 was set for the address (which is overridden by the transport).
3480 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3483 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3484 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3485 because it is too late at that time, and has no effect.
3487 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3488 security issue with \' (bugzilla #107). I could not use the
3489 PQescapeStringConn() function, because it needs a PGconn value as one of
3492 PH/08 When testing addresses using -bt, indicate those final addresses that
3493 are duplicates that would not cause an additional delivery. At least one
3494 person was confused, thinking that -bt output corresponded to deliveries.
3495 (Suppressing duplicates isn't a good idea as you lose the information
3496 about possibly different redirections that led to the duplicates.)
3498 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3499 systems where poll() doesn't work, in particular OS X.
3501 PH/10 Added more information to debugging output for retry time not reached.
3503 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3504 operations in malware.c.
3506 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3509 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3510 syslog_duplication set false, log lines that would normally be written
3511 both the the main log and to the reject log were not written to syslog at
3514 PH/14 In the default configuration, change the use of "message" in ACL warn
3515 statements to "add_header".
3517 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3518 not followed by a command (e.g. "seen endif").
3520 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3521 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3524 PH/17 Added extra conditions to the default value of delay_warning_condition
3528 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3529 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3530 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3533 The Auto-Submitted: and various List- headers are standardised, whereas I
3534 don't think Precedence: ever was.
3536 PH/18 Refactored debugging code in route_finduser() to show more information,
3537 in particular, the error code if getpwnam() issues one.
3539 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3540 This is apparently needed in addition to the PH/07 change above to avoid
3541 any possible encoding problems.
3543 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3544 but not after initializing Perl.
3546 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3547 output them only if debugging. By default they are written stderr,
3548 apparently, which is not desirable.
3550 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3553 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3556 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3558 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3559 authenticated or an ident call has been made. Suppress the default
3560 values for $authenticated_id and $authenticated_sender (but permit -oMai
3561 and -oMas) when testing with -bh.
3563 PH/24 Re-jigged the order of the tests in the default configuration so that the
3564 tests for valid domains and recipients precede the DNS black list and CSA
3565 tests, on the grounds that those ones are more expensive.
3567 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3568 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3569 This bug exists in every version of Exim that I still have, right back to
3572 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3573 However, an attempt to turn on case-sensitivity in a regex key by
3574 including (?-i) didn't work because the subject string was already
3575 lowercased, and the effects were non-intuitive. It turns out that a
3576 one-line patch can be used to allow (?-i) to work as expected.
3582 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3583 other effects) broke the use of negated acl sub-conditions.
3585 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3588 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3589 "Deny" causes Exim to reject the incoming connection with a 554 error.
3590 Unfortunately, if there is a major crisis, such as a disk failure,
3591 tcp-wrappers gives "deny", whereas what one would like would be some
3592 kind of temporary error. A kludge has been added to help with this.
3593 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3594 554 error is used if errno is still zero or contains ENOENT (which occurs
3595 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3598 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3600 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3601 errors. Otherwise a message that provokes a temporary error (when other
3602 messages do not) can cause a whole host to time out.
3604 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3605 addresses were routed directly to files or pipes from a redirect router.
3606 File deliveries just didn't batch; pipe deliveries might have suffered
3609 PH/06 A failure to get a lock for a hints database would erroneously always say
3610 "Failed to get write lock", even when it was really a read lock.
3612 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3613 of 0600. This has been changed to use the value of the lockfile_mode
3614 option (which defaults to 0600).
3616 PH/08 Applied small patch from the Sieve maintainer.
3618 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3619 folder from quota calculations, a direct delivery into this folder messed
3620 up the contents of the maildirsize file. This was because the regex was
3621 used only to exclude .Trash (or whatever) when the size of the mailbox
3622 was calculated. There was no check that a delivery was happening into an
3623 excluded directory. This bug has been fixed by ignoring all quota
3624 processing for deliveries into excluded directories.
3626 PH/10 Added the maildirfolder_create_regex option to appendfile.
3632 PH/01 The code for finding all the local interface addresses on a FreeBSD
3633 system running IPv6 was broken. This may well have applied to all BSD
3634 systems, as well as to others that have similar system calls. The broken
3635 code found IPv4 interfaces correctly, but gave incorrect values for the
3636 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3637 that it would not match correctly against @[] and not recognize the IPv6
3640 PH/02 The ipliteral router was not recognizing addresses of the form user@
3641 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3643 PH/03 Added disable_ipv6.
3645 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3646 decoded form, because it is most often used to construct To: headers
3647 lines in autoreplies, and the decoded form may well be syntactically
3648 invalid. However, $reply_address has leading white space removed, and all
3649 newlines turned into spaces so that the autoreply transport does not
3652 PH/05 If group was specified without a user on a router, and no group or user
3653 was specified on a transport, the group from the router was ignored.
3655 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3656 for visible compile-time settings that can be used to change these
3657 numbers, for those that want even more. Backwards compatibility with old
3658 spool files has been maintained. However, going back to a previous Exim
3659 release will lost any variables that are in spool files.
3661 PH/07 Two small changes when running in the test harness: increase delay when
3662 passing a TCP/IP connection to a new process, in case the original
3663 process has to generate a bounce, and remove special handling of
3664 127.0.0.2 (sic), which is no longer necessary.
3666 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3667 be the same on different OS.
3669 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3672 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3673 whether --show-vars was specified or not
3675 JJ/02 exipick: Added support for new ACL variable spool format introduced
3678 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3679 syntactically invalid From: or Reply-to: line, and a filter used this to
3680 generate an autoreply, and therefore failed to obtain an address for the
3681 autoreply, Exim could try to deliver to a non-existent relative file
3682 name, causing unrelated and misleading errors. What now happens is that
3683 it logs this as a hard delivery error, but does not attempt to create a
3686 PH/11 The exinext utility has a -C option for testing purposes, but although
3687 the given file was scanned by exinext itself; it wasn't being passed on
3688 when Exim was called.
3690 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3691 an end-of-file indication when reading a command response.
3693 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3694 compiled. In many other places in Exim, IPv6 addresses are always
3695 recognized, so I have changed this. It also means that IPv4 domain
3696 literals of the form [IPV4:n.n.n.n] are now always recognized.
3698 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3699 used if the router is not running as root, for example, when verifying at
3700 ACL time, or when using -bh. The debugging output from this situation was
3701 non-existent - all you got was a failure to exec. I have made two
3704 (a) Failures to set uid/gid, the current directory, or a process leader
3705 in a subprocess such as that created by queryprogram now generate
3706 suitable debugging output when -d is set.
3708 (b) The queryprogram router detects when it is not running as root,
3709 outputs suitable debugging information if -d is set, and then runs
3710 the subprocess without attempting to change uid/gid.
3712 PH/15 Minor change to Makefile for building test_host (undocumented testing
3715 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3716 additional section of a DNS packet that returns MX or SRV records.
3717 Instead, it always explicitly searches for A/AAAA records. This avoids
3718 major problems that occur when a DNS server includes only records of one
3719 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3720 fixed another bug: if SRV records were looked up and the corresponding
3721 address records were *not* found in the additional section, the port
3722 values from the SRV records were lost.
3724 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3725 using the correct key (the original address) when searching the retry
3726 rules in order to find which one to use for generating the retry hint.
3728 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3729 adding the default one. Similarly, if it contains a Reply-To: header, the
3730 errors_reply_to option, if set, is not used.
3732 PH/19 When calculating a retry time, Exim used to measure the "time since
3733 failure" by looking at the "first failed" field in the retry record. Now
3734 it does not use this if it is later than than the arrival time of the
3735 message. Instead it uses the arrival time. This makes for better
3736 behaviour in cases where some deliveries succeed, thus re-setting the
3737 "first failed" field. An example is a quota failure for a huge message
3738 when small messages continue to be delivered. Without this change, the
3739 "time since failure" will always be short, possible causing more frequent
3740 delivery attempts for the huge message than are intended.
3741 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3743 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3744 $1, $2, $3) because the numerical variables can be reset during some
3745 expansion items (e.g. "match"), thereby losing the authentication data.
3747 PH/21 Make -bV show the size of off_t variables so that the test suite can
3748 decide whether to run tests for quotas > 2G.
3750 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3751 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3752 filecount value is greater than 2G or if a quota value is greater than 2G
3753 on a system where the size of off_t is not greater than 4, a panic error
3756 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3757 never match. The debug and -bh output now contains an explicit error
3758 message indicating a malformed IPv4 address or mask.
3760 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3761 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3762 PH/23 above applies.
3764 PH/25 Do not write to syslog when running in the test harness. The only
3765 occasion when this arises is a failure to open the main or panic logs
3766 (for which there is an explicit test).
3768 PH/26 Added the /no_tell option to "control=freeze".
3770 PH/27 If a host name lookup failed very early in a connection, for example, if
3771 the IP address matched host_lookup and the reverse lookup yielded a name
3772 that did not have a forward lookup, an error message of the form "no IP
3773 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3774 could be logged. Now it outputs the IP address instead of "NULL".
3776 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3777 allows the sender and the authenticated sender to be set when
3778 submitting a message from within Exim. Since child_open_exim() is
3779 documented for local_scan(), the new function should be too.
3781 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3782 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3783 results in an empty string is now treated as unset.
3785 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3787 PH/31 Added sender_verify_fail logging option.
3789 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3790 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3791 tidied the source and removed it altogether.
3793 PH/33 When a queue run was abandoned because the load average was too high, a
3794 log line was always written; now it is written only if the queue_run log
3795 selector is set. In addition, the log line for abandonment now contains
3796 information about the queue run such as the pid. This is always present
3797 in "start" and "stop" lines but was omitted from the "abandon" line.
3799 PH/34 Omit spaces between a header name and the colon in the error message that
3800 is given when verify = headers_syntax fails (if there are lots of them,
3801 the message gets confusing).
3803 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3804 names, as there are now some PTR records that contain slashes. This check
3805 is only to protect against broken name servers that fall over on strange
3806 characters, so the fact that it applies to all lookups doesn't matter.
3808 PH/36 Now that the new test suite is complete, we can remove some of the
3809 special code in Exim that was needed for the old test suite. For example,
3810 sorting DNS records because real resolvers return them in an arbitrary
3811 order. The new test suite's fake resolver always returns records in the
3814 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3815 bounces) except when queue_only is set, to avoid logging races between
3816 the different processes.
3818 PH/38 Panic-die if .include specifies a non-absolute path.
3820 PH/39 A tweak to the "H" retry rule from its user.
3822 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3823 a label. They prevented compilation on older perls.
3825 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3826 a warning to be raised on newish perls.
3828 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3829 on queue. Changes to match documented behaviour of showing count of
3830 messages matching specified criteria.
3832 PH/40 Changed the default ident timeout from 30s to 5s.
3834 PH/41 Added support for the use of login_cap features, on those BSD systems
3835 that have them, for controlling the resources used by pipe deliveries.
3837 PH/42 The content-scanning code uses fopen() to create files in which to put
3838 message data. Previously it was not paying any attention to the mode of
3839 the files. Exim runs with umask(0) because the rest of the code creates
3840 files with open(), and sets the required mode explicitly. Thus, these
3841 files were ending up world-writeable. This was not a big issue, because,
3842 being within the spool directory, they were not world-accessible. I have
3843 created a function called modefopen, which takes an additional mode
3844 argument. It sets umask(777), creates the file, chmods it to the required
3845 mode, then resets the umask. All the relevant calls to fopen() in the
3846 content scanning code have been changed to use this function.
3848 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3849 to 24 hours. This avoids potential overflow problems when processing G
3850 and H retry rules. I suspect nobody ever tinkers with this value.
3852 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3854 PH/45 When the plaintext authenticator is running as a client, the server's
3855 challenges are checked to ensure they are valid base64 strings. By
3856 default, the authentication attempt is cancelled if an invalid string is
3857 received. Setting client_ignore_invalid_base64 true ignores these errors.
3858 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3859 they are received. Thus, the responses can be made to depend on the
3860 challenges. If an invalid string is ignored, an empty string is placed in
3863 PH/46 Messages that are created by the autoreply transport now contains a
3864 References: header, in accordance with RFCs 2822 and 3834.
3866 PH/47 Added authenticated_sender_force to the smtp transport.
3868 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3870 PH/49 Installed latest patch from the Sieve maintainer.
3872 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3873 was also set, the appendfile transport was unnecessarily scanning a
3874 directory of message files (e.g. for maildir delivery) to find the count
3875 of files (along with the size), even though it did not need this
3876 information. It now does the scan only if it needs to find either the
3877 size of the count of files.
3879 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3881 PH/52 Two bugs concerned with error handling when the smtp transport is
3884 (i) Exim was not creating retry information for temporary errors given
3885 for individual recipients after the DATA command when the smtp transport
3886 was used in LMTP mode. This meant that they could be retried too
3887 frequently, and not timed out correctly.
3889 (ii) Exim was setting the flag that allows error details to be returned
3890 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3891 recipients that were returned after the DATA command.
3893 PH/53 This is related to PH/52, but is more general: for any failing address,
3894 when detailed error information was permitted to be returned to the
3895 sender, but the error was temporary, then after the final timeout, only
3896 "retry timeout exceeded" was returned. Now it returns the full error as
3897 well as "retry timeout exceeded".
3899 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3900 do this, and (what is worse) MTAs that accept it.
3902 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3903 will now be deprecated.
3905 PH/56 New os.c-cygwin from the Cygwin maintainer.
3907 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3908 formats (previously only available in exim formats via -bpr, -bpru,
3909 and -bpra. Now also available in native and exiqgrep formats)
3911 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3912 with very large, slow to parse queues
3914 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3916 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3918 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3919 responses to authentication challenges, though it was showing the
3920 challenges; (ii) I've removed the CR characters from the debug output for
3923 PH/58 Allow for the insertion of a newline as well as a space when a string
3924 is turned into more than one encoded-word during RFC 2047 encoding. The
3925 Sieve code now uses this.
3927 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3928 data_4xx, lost_connection, tls_required.
3930 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3931 message was being sent as an SMTP response.
3933 PH/61 Add -l and -k options to exicyclog.
3935 PH/62 When verifying, if an address was redirected to one new address, so that
3936 verification continued, and the new address failed or deferred after
3937 having set something in $address_data, the value of $address_data was not
3938 passed back to the ACL. This was different to the case when no
3939 redirection occurred. The value is now passed back in both cases.
3941 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3942 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3943 use login_cap.h, so on its own it isn't the distinguishing feature. The
3944 new name refers directly to the setclassresources() function.
3946 PH/65 Added configuration files for NetBSD3.
3948 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3950 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3951 is preferred over IPv4.
3953 PH/68 The bounce_return_message and bounce_return_body options were not being
3954 honoured for bounces generated during the reception of non-SMTP messages.
3955 In particular, this applied to messages rejected by the ACL. This bug has
3956 been fixed. However, if bounce_return_message is true and bounce_return_
3957 body is false, the headers that are returned for a non-SMTP message
3958 include only those that have been read before the error was detected.
3959 (In the case of an ACL rejection, they have all been read.)
3961 PH/69 The HTML version of the specification is now built in a directory called
3962 spec_html instead of spec.html, because the latter looks like a path with
3963 a MIME-type, and this confuses some software.
3965 PH/70 Catch two compiler warnings in sieve.c.
3967 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3968 function verify_get_ident() calls ip_connect() to connect a socket, but
3969 if the "connect()" function timed out, ip_connect() used to close the
3970 socket. However, verify_get_ident() also closes the socket later, and in
3971 between Exim writes to the log, which may get opened at this point. When
3972 the socket was closed in ip_connect(), the log could get the same file
3973 descriptor number as the socket. This naturally causes chaos. The fix is
3974 not to close the socket in ip_connect(); the socket should be closed by
3975 the function that creates it. There was only one place in the code where
3976 this was missing, in the iplookup router, which I don't think anybody now
3977 uses, but I've fixed it anyway.
3979 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3980 well as to direct DNS lookups. Otherwise the handling of names in host
3981 lists is inconsistent and therefore confusing.
3987 PH/01 Two changes to the default runtime configuration:
3989 (1) Move the checks for relay_from_hosts and authenticated clients from
3990 after to before the (commented out) DNS black list checks.
3992 (2) Add control=submission to the relay_from_hosts and authenticated
3993 clients checks, on the grounds that messages accepted by these
3994 statements are most likely to be submissions.
3996 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3998 (1) Generate an error if the third argument for the ${prvs expansion is
4001 (2) Treat a missing third argument of ${prvscheck as if it were an empty
4004 (3) Reset the variables that are obtained from the first argument of
4005 ${prvscheck and used in the second argument before leaving the code,
4006 because their memory is reclaimed, so using them afterwards may do
4009 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
4010 one (it's much easier than Tom thought :-).
4012 (5) Because of (4), we can now allow for the use of $prvscheck_result
4013 inside the third argument.
4015 PH/03 For some reason, the default setting of PATH when running a command from
4016 a pipe transport was just "/usr/bin". I have changed it to
4019 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
4020 anything to be listed in the output from -bV.
4022 PH/05 When a filter generated an autoreply, the entire To: header line was
4023 quoted in the delivery log line, like this:
4025 => >A.N.Other <ano@some.domain> <original@ddress> ...
4027 This has been changed so that it extracts the operative address. There
4028 may be more than one such address. If so, they are comma-separated, like
4031 => >ano@some.domain,ona@other.domain <original@ddress> ...
4033 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
4034 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
4035 not being looked up in the rDNS to get a host name, Exim was showing the
4036 IP address twice in Received: lines, even though the IP addresses were
4037 identical. For example:
4039 Received: from [1.2.3.4] (helo=[1.2.3.4])
4041 However, if the real host name was known, it was omitting the HELO data
4042 if it matched the actual IP address. This has been tidied up so that it
4043 doesn't show the same IP address twice.
4045 PH/07 When both +timestamp and +memory debugging was on, the value given by
4046 $tod_xxx expansions could be wrong, because the tod_stamp() function was
4047 called by the debug printing, thereby overwriting the timestamp buffer.
4048 Debugging no longer uses the tod_stamp() function when +timestamp is set.
4050 PH/08 When the original message was included in an autoreply transport, it
4051 always said "this is a copy of the message, including all the headers",
4052 even if body_only or headers_only was set. It now gives an appropriate
4055 PH/09 Applied a patch from the Sieve maintainer which:
4057 o fixes some comments
4058 o adds the (disabled) notify extension core
4059 o adds some debug output for the result of if/elsif tests
4060 o points to the current vacation draft in the documentation
4061 and documents the missing references header update
4065 o fixes a bug in processing the envelope test (when testing
4066 multiple envelope elements, the last element determined the
4069 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
4070 Electronic Mail") by including:
4072 Auto-submitted: auto-generated
4074 in the messages that it generates (bounce messages and others, such as
4075 warnings). In the case of bounce messages for non-SMTP messages, there was
4076 also a typo: it was using "Auto_submitted" (underscore instead of
4077 hyphen). Since every message generated by Exim is necessarily in response
4078 to another message, thes have all been changed to:
4080 Auto-Submitted: auto-replied
4082 in accordance with these statements in the RFC:
4084 The auto-replied keyword:
4086 - SHOULD be used on messages sent in direct response to another
4087 message by an automatic process,
4089 - MUST NOT be used on manually-generated messages,
4091 - MAY be used on Delivery Status Notifications (DSNs) and Message
4092 Disposition Notifications (MDNs),
4094 - MUST NOT be used on messages generated by automatic or periodic
4095 processes, except for messages which are automatic responses to
4098 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
4099 to the default Received: header definition.
4101 PH/12 Added log selector acl_warn_skipped (default on).
4103 PH/13 After a successful wildlsearch lookup, discard the values of numeric
4104 variables because (a) they are in the wrong storage pool and (b) even if
4105 they were copied, it wouldn't work properly because of the caching.
4107 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
4108 checking when decoding. Apparently there are clients that generate
4109 overlong encoded strings. Why am I not surprised?
4111 PH/15 If the first argument of "${if match_address" was not empty, but did not
4112 contain an "@" character, Exim crashed. Now it writes a panic log message
4113 and treats the condition as false.
4115 PH/16 In autoreply, treat an empty string for "once" the same as unset.
4117 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
4118 extension "envelope-auth". The code is finished and in agreement with
4119 other implementations, but there is no documentation so far and in fact,
4120 nobody wrote the draft yet. This extension is currently #undef'ed, thus
4121 not changing the active code.
4123 Print executed "if" and "elsif" statements when debugging is used. This
4124 helps a great deal to understand what a filter does.
4126 Document more things not specified clearly in RFC3028. I had all this
4127 sorted out, when out of a sudden new issues came to my mind. Oops."
4129 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
4132 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
4133 canonical form (as documented). However, after a host name lookup from
4134 the IP address, check_host() was doing a simple string comparison with
4135 addresses acquired from the DNS when checking that the found name did
4136 have the original IP as one of its addresses. Since any found IPv6
4137 addresses are likely to be in abbreviated form, the comparison could
4138 fail. Luckily, there already exists a function for doing the comparison
4139 by converting both addresses to binary, so now that is used instead of
4140 the text comparison.
4142 PH/20 There was another similar case to PH/19, when a complete host name was
4143 given in a host list; looking up its IP address could give an abbreviated
4144 form, whereas the current host's name might or might not be abbreviated.
4145 The same fix has been applied.
4151 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
4152 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
4155 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
4156 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
4158 PH/03 Typo: missing ".o" in src/pcre/Makefile.
4160 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
4161 header line, restrict the check to what is listed in RFCs 2369 and 2929.
4162 Also, for "Auto-Submitted", treat anything other than "no" as
4163 non-personal, in accordance with RFC 3834. (Previously it treated
4164 anything starting "auto-" as non-personal.)
4166 TF/01 The control=submission/name=... option had a problem with syntax
4167 errors if the name included a slash character. The /name= option
4168 now slurps the rest of the string, so it can include any characters
4169 but it must come last in the list of options (after /sender_retain
4172 PH/05 Some modifications to the interface to the fake nameserver for the new
4180 TK/01 Added the "success_on_redirect" address verification option. See
4181 NewStuff for rationale and an example.
4183 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
4185 PH/02 Patch to exigrep to allow it to work on syslog lines.
4187 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
4188 fread() to skip over the body file's header line, because in Cygwin the
4189 header line is locked and is inaccessible.
4191 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
4192 co-exist for some time) to make it clear that it is the Exim ID that is
4193 referenced, not the Message-ID: header line.
4195 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
4196 string_format() function, because snprintf() does not exist on all
4199 PH/06 The use of forbid_filter_existstest now also locks out the use of the
4200 ${stat: expansion item.
4202 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
4203 protocol synchronization error", to keep the pedants happy.
4205 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
4206 well as for IRIX systems, when gcc is being used. See the host.c source
4209 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
4211 PH/10 Named domain lists were not working if used in a queue_smtp_domains
4214 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
4215 transport and to the smtp transport in LMTP mode.
4217 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
4219 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
4220 run a filter in a subprocess. This could lead to confusion in subsequent
4221 lookups in the parent process. There should also be a search_tidyup() at
4222 the end of the subprocess.
4224 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
4225 only if the host matched helo_try_verify_hosts, which caused the
4226 verification to occur when the EHLO/HELO command was issued. The ACL just
4227 tested the remembered result. Now, if a previous verification attempt has
4228 not happened, "verify = helo" does it there and then.
4230 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
4232 TK/03 Fix log output including CR from clamd.
4234 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
4235 exist provoked a memory error which could cause a segfault.
4237 PH/15 Installed PCRE 6.2
4239 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
4241 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
4242 of the problem. Specifically, suggested +O2 rather than +O1 for the
4245 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
4247 PH/20 If a delivery was routed to a non-standard port by means of an SRV
4248 record, the port was not correctly logged when the outgoing_port log
4249 selector was set (it logged the transort's default port).
4251 PH/21 Added support for host-specific ports to manualroute, queryprogram,
4252 fallback_hosts, and "hosts" in the smtp transport.
4254 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
4255 host errors such as "Connection refused".
4257 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
4258 authentication with radiusclient 0.4.9:
4260 - Error returned from rc_read_config was caught wrongly
4261 - Username/password not passed on to radius server due to wrong length.
4263 The presumption is that some radiusclient API changes for 4.51/PH/17
4264 were not taken care of correctly. The code is still untested by me (my
4265 Linux distribution still has 0.3.2 of radiusclient), but it was
4266 contributed by a Radius user.
4268 PH/24 When doing a callout, the value of $domain wasn't set correctly when
4269 expanding the "port" option of the smtp transport.
4271 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
4272 while reading a MIME header. Thanks to Tom Hughes for a patch.
4274 PH/24 Include config.h inside local_scan.h so that configuration settings are
4277 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
4278 that in an ACL for RCPT (for example), you can examine exactly what was
4281 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
4282 commands, but it was not correctly comparing the address with the actual
4283 client host address. Thus, it would show the EHLO address in Received:
4284 header lines when this was not necessary.
4286 PH/27 Added the % operator to ${eval:}.
4288 PH/28 Exim tries to create and chdir to its spool directory when it starts;
4289 it should be ignoring failures (because with -C, for example, it has lost
4290 privilege). It wasn't ignoring creation failures other than "already
4293 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
4296 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
4297 because an input line was too long, either on its own, or by virtue of
4298 too many continuations, the temporary file was not being removed, and the
4299 return code was incorrect.
4301 PH/31 Missing "BOOL" in function definition in filtertest.c.
4303 PH/32 Applied Sieve patches from the maintainer.
4305 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
4307 PH/33 Added "verify = not_blind".
4309 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
4310 Local/Makefile (with some defaults set). These are used in built scripts
4311 such as exicyclog, but they have never been used in the exim_install
4312 script (though there are many overriding facilities there). I have
4313 arranged that the exim_install script now takes note of these two
4316 PH/35 Installed configuration files for Dragonfly.
4318 PH/36 When a locally submitted message by a trusted user did not contain a
4319 From: header, and the sender address was obtained from -f or from an SMTP
4320 MAIL command, and the trusted user did not use -F to supply a sender
4321 name, $originator_name was incorrectly used when constructing a From:
4322 header. Furthermore, $originator_name was used for submission mode
4323 messages from external hosts without From: headers in a similar way,
4324 which is clearly wrong.
4326 PH/37 Added control=suppress_local_fixups.
4328 PH/38 When log_selector = +received_sender was set, and the addition of the
4329 sender made the log line's construction buffer exactly full, or one byte
4330 less than full, an overflow happened when the terminating "\n" was
4333 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
4334 when the result of a list match is failure because a DNS lookup failed.
4336 PH/40 RM_COMMAND is now used in the building process.
4338 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
4339 the "build-* directories that it finds.
4341 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
4342 address in a domain literal was a prefix of an interface address.
4344 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
4345 when verifying a sender address, unless rewrite_headers is false.
4347 PH/44 Wrote a long comment about why errors_to addresses are verified as
4348 recipients, not senders.
4350 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
4351 the ratelimit ACL was added.
4353 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
4355 PH/46 Added extra information about PostgreSQL errors to the error string.
4357 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
4358 avoiding the need to install special zones in a real server. This is
4359 backwards compatible; if it can't find the fake resolver, it drops back.
4360 Thus, both old and new test suites can be run.
4362 TF/02 Added util/ratelimit.pl
4364 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
4365 clock is set back in time.
4367 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
4368 Candler <B.Candler@pobox.com>.
4370 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
4371 OK for addresses that are the result of redirections.
4373 PH/48 A number of further additions for the benefit of the new test suite,
4374 including a fake gethostbyname() that interfaces to the fake DNS resolver
4377 TF/06 The fix for widen_domains has also been applied to qualify_single and
4378 search_parents which are the other dnslookup options that can cause
4381 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
4384 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
4386 TF/07 Exim produced the error message "an SRV record indicated no SMTP
4387 service" if it encountered an MX record with an empty target hostname.
4388 The message is now "an MX or SRV record indicated no SMTP service".
4390 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
4391 if the DNS of the sending site is misconfigured. This is quite a
4392 common situation. This change restores the behaviour of treating a
4393 helo verification defer as a failure.
4395 PH/51 If self=fail was set on a router, the bounce message did not include the
4396 actual error message.
4402 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
4404 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
4405 command itself ended in error, the underlying message about the transport
4406 filter timeout was being overwritten with the pipe command error. Now the
4407 underlying error message should be appended to the second error message.
4409 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
4411 PH/02 Reduce the amount of output that "make" produces by default. Full output
4412 can still be requested.
4414 PH/03 The warning log line about a condition test deferring for a "warn" verb
4415 was being output only once per connection, rather than after each
4416 occurrence (because it was using the same function as for successful
4417 "warn" verbs). This seems wrong, so I have changed it.
4419 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
4420 it should not have, which might have caused a crash in the right
4421 circumstances, but probably never did.
4423 PH/04 Installed a modified version of Tony Finch's patch to make submission
4424 mode fix the return path as well as the Sender: header line, and to
4425 add a /name= option so that you can make the user's friendly name appear
4428 TF/03 Added the control = fakedefer ACL modifier.
4430 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
4431 Mark Lowes for thorough testing.
4433 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
4435 TK/03 Merged latest SRS patch from Miles Wilton.
4437 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
4438 with the definition in sysexits.h (which is #included earlier).
4439 Fortunately, Exim does not actually use EX_OK. The code used to try to
4440 preserve the sysexits.h value, by assuming that macro definitions were
4441 scanned for macro replacements. I have been disabused of this notion,
4442 so now the code just undefines EX_OK before #including unistd.h.
4444 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
4445 in the smtp transport. When a block could not be written in a single
4446 write() function, the timeout was being re-applied to each part-write.
4447 This seems wrong - if the receiver was accepting one byte at a time it
4448 would take for ever. The timeout is now adjusted when this happens. It
4449 doesn't have to be particularly precise.
4451 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
4452 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
4454 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
4455 without a domain if the check to <postmaster@domain> fails.
4457 SC/01 Eximstats: added -xls and the ability to specify output files
4458 (patch written by Frank Heydlauf).
4460 SC/02 Eximstats: use FileHandles for outputting results.
4462 SC/03 Eximstats: allow any combination of xls, txt, and html output.
4464 SC/04 Eximstats: fixed display of large numbers with -nvr option
4466 SC/05 Eximstats: fixed merging of reports with empty tables.
4468 SC/06 Eximstats: added the -include_original_destination flag
4470 SC/07 Eximstats: removed tabs and trailing whitespace.
4472 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
4474 TK/06 MBOX spool code: Add real "From " MBOX separator line
4475 so the .eml file is really in mbox format (even though
4476 most programs do not really care). Patch from Alex Miller.
4478 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4479 The latter is generated from $received_to and is only set if the
4480 message has one envelope recipient. SA can use these headers,
4481 obviously out-of-the-box. Patch from Alex Miller.
4483 PH/08 The ${def test on a variable was returning false if the variable's
4484 value was "0", contrary to what the specification has always said!
4485 The result should be true unless the variable is empty.
4487 PH/09 The syntax error of a character other than { following "${if
4488 def:variable_name" (after optional whitespace) was not being diagnosed.
4489 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4490 accidental colon was present, for example, could give incorrect results.
4492 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4493 result is used (not including appendfile, where other changes are about
4496 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4497 This involved changing a lot of size variables from int to off_t. It
4498 should work with maildirs and everything.
4500 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4501 spamd dying while we are connected to it.
4503 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4506 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4507 function for BDB 4.3.
4509 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4511 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4512 the sources for PCRE have been re-arranged and more files are now
4515 PH/15 The code I had for printing potentially long long variables in PH/11
4516 above was not the best (it lost precision). The length of off_t variables
4517 is now inspected at build time, and an appropriate printing format (%ld
4518 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4519 to be "long long int" or "long int". This is needed for the internal
4520 formatting function string_vformat().
4522 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4523 the configuration file to be ":syslog", then the script "guesses" where
4524 the logs files are, rather than using the compiled in default. In our
4525 case the guess is not the same as the compiled default, so the script
4526 suddenly stopped working when I started to use syslog. The patch checks
4527 to see if log_file_path is "". If so, it attempts to read it from exim
4528 with no configuration file to get the compiled in version, before it
4529 falls back to the previous guessing code."
4531 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4532 implementing BATV in an Exim configuration. See NewStuff for the gory
4535 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4536 Makefile that are specific to HP-UX.
4538 PH/18 If the "use_postmaster" option was set for a recipient callout together
4539 with the "random" option, the postmaster address was used as the MAIL
4540 FROM address for the random test, but not for the subsequent recipient
4541 test. It is now used for both.
4543 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4544 patch removes a few documentation additions to RFC 3028, because the
4545 latest draft now contains them. It adds the new en;ascii-case comparator
4546 and a new error check for 8bit text in MIME parts. Comparator and
4547 require names are now matched exactly. I enabled the subaddress
4548 extension, but it is not well tested yet (read: it works for me)."
4550 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4551 rework some of the code of TK/09 above to avoid the hardwired use of
4552 "%lld" and "long long". Replaced the call to snprintf() with a call to
4555 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4556 records point to non-existent hosts", "retry timeout exceeded", and
4557 "retry time not reached for any host after a long failure period".
4559 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4560 experimental DomainKeys support:
4562 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4563 (2) On an error such as an illegally used "control", the wrong name for
4564 the control was given.
4566 These problems did NOT occur unless DomainKeys support was compiled.
4568 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4570 PH/24 Added ${if match_ip condition.
4572 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4573 databases so that it will be absolutely obvious if a crash occurs in the
4574 DB library. This is a regular occurrence (often caused by mis-matched
4577 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4578 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4579 picked up on a user's system that detects such things. There doesn't seem
4580 to be a gcc warning option for this - only an attribute that has to be
4581 put on the function's prototype. It seems that in Fedora Core 4 they have
4582 set this on a number of new functions. No doubt there will be more in due
4585 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4586 specify a transport. However, if an address that was verified by such a
4587 router was the subject of a callout, Exim crashed because it tried to
4588 read the rcpt_include_affixes from the non-existent transport. Now it
4589 just assumes that the setting of that option is false. This bug was
4590 introduced by 4.51/PH/31.
4592 PH/28 Changed -d+all to exclude +memory, because that information is very
4593 rarely of interest, but it makes the output a lot bigger. People tend to
4594 do -d+all out of habit.
4596 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4597 code in os-type was giving problems when libc.so lives in lib64, like on
4600 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4601 aren't the modern standard, and it seems that some systems' include files
4602 don't always have them. Exim was already checking for some of the newer
4603 ones like T_AAAA, and defining it itself. I've added checks for all the
4604 record types that Exim uses.
4606 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4607 not automatically generating a new one, as it is supposed to. This
4608 prevented TLS from working. If the file did exist, but contained invalid
4609 data, a new version was generated, as expected. It was only the case of a
4610 non-existent file that was broken.
4612 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4613 with a change in libdomainkeys > 0.64.
4615 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4616 from DNS. If the selector record carries the flag, it now has
4617 precedence over the domain-wide flag.
4619 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4621 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4622 the use of a port name that isn't defined in /etc/services) occurred, the
4623 message was deferred as in a normal delivery, and thus remained on the
4624 spool, instead of being failed because of the mua_wrapper setting. This
4625 is now fixed, and I tidied up some of the mua_wrapper messages at the
4628 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4629 the messages in a hash of arrays rather than using individual hashes.
4630 This is a bit cleaner and results in dramatic memory savings, albeit
4631 at a slight CPU cost.
4633 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4634 as requested by Marc Sherman.
4636 SC/10 Eximstats: added histograms for user specified patterns as requested
4639 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4641 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4642 fopen() in the content-scanning modules that did not already have it.
4648 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4649 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4651 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4653 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4655 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4656 to test Sieve filters that use "vacation".
4658 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4659 that changes the way the GnuTLS parameters are stored in the cache file.
4660 The new format can be generated externally. For backward compatibility,
4661 if the data in the cache doesn't make sense, Exim assumes it has read an
4662 old-format file, and it generates new data and writes a new file. This
4663 means that you can't go back to an older release without removing the
4666 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4667 work if there are any delivery delays because "one_time" forces the
4668 parent to be marked "delivered", so its unseen clone is never tried
4669 again. For this reason, Exim now forbids the simultaneous setting of
4672 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4673 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4674 entry if you want to know the details. The fix, however, neglected to
4675 consider the case when local delivery batching is involved. The test for
4676 "previously delivered" was not happening when checking to see if an
4677 address could be batched with a previous (undelivered) one; under
4678 certain circumstances this could lead to multiple deliveries to the same
4681 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4682 in its include files, and this causes problems building Exim.
4684 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4685 header_syntax) but Exim was just ignoring anything given after a slash.
4686 In particular, this caused confusion with an attempt to use "verify =
4687 reverse_host_lookup/defer_ok". An error is now given when options are
4688 supplied for verify items that do not have them. (Maybe reverse_host_
4689 lookup should have a defer_ok option, but that's a different point.)
4691 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4692 defined by RFC 821) to 2048, because there were problems with some AUTH
4693 commands, and RFC 1869 says the size should be increased for extended
4694 SMTP commands that take arguments.
4696 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4699 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4700 "unknown" error; now it says that the functionality isn't in the binary.
4702 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4703 an address' error message when a string expansion fails (syntax or
4704 whatever). Otherwise the password may appear in the log. Following change
4705 PH/42 below, there is no longer a chance of it appearing in a bounce
4708 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4710 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4711 of its addresses was ever tried. (Bugzilla bug #2.)
4713 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4714 the result incorrectly in the debug output. (It correctly added a newline
4715 to what was transported.)
4717 TF/01 Added $received_time.
4719 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4720 commented out examples of how to interface to a virus scanner and to
4721 SpamAssassin. Also added commented examples of av_scanner and
4722 spamd_address settings.
4724 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4725 and controls are allowed in which ACLs. There were a couple of minor
4726 errors. Some of the entries in the conditions table (which is a table of
4727 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4728 negation of where the condition IS allowed.
4730 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4732 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4733 header file does not have a version number, so I've had to invent a new
4734 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4735 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4736 radiusclient), but it was contributed by a Radius user.
4738 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4739 files or directories, for OpenSSL.
4741 PH/19 When an Exim process that is running as root has to create an Exim log
4742 file, it does so in a subprocess that runs as exim:exim so as to get the
4743 ownership right at creation (otherwise, other Exim processes might see
4744 the file with the wrong ownership). There was no test for failure of this
4745 fork() call, which would lead to the process getting stuck as it waited
4746 for a non-existent subprocess. Forks do occasionally fail when resources
4747 run out. I reviewed all the other calls to fork(); they all seem to check
4750 PH/20 When checking for unexpected SMTP input at connect time (before writing
4751 the banner), Exim was not dealing correctly with a non-positive return
4752 from the read() function. If the client had disconnected by this time,
4753 the result was a log entry for a synchronization error with an empty
4754 string after "input=" when read() returned zero. If read() returned -1
4755 (an event I could not check), uninitialized data bytes were printed.
4756 There were reports of junk text (parts of files, etc) appearing after
4759 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4761 PH/22 Added support for macro redefinition, and (re)definition in between
4762 driver and ACL definitions.
4764 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4765 forgetting to use the resulting value; it was using the unexpanded value.
4767 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4768 hadn't been configured. The fix is from Juergen Kreileder, who
4769 understands it better than I do:
4771 "Here's what I see happening with three configured cyrus_sasl
4772 authenticators configured (plain, login, cram-md5):
4774 On startup auth_cyrus_sasl_init() gets called for each of these.
4775 This means three calls to sasl_listmech() without a specified mech_list.
4776 => SASL tests which mechs of all available mechs actually work
4777 => three warnings about OTP not working
4778 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4780 With the patch, sasl_listmech() also gets called three times. But now
4781 SASL's mech_list option is set to the server_mech specified in the the
4782 authenticator. Or in other words, the answer from sasl_listmech()
4783 gets limited to just the mech you're testing for (which is different
4785 => the return list contains just 'plain' or 'login', 'cram-md5' or
4786 nothing depending on the value of ob->server_mech.
4788 I've just tested the patch: Authentication still works fine,
4789 unavailable mechs specified in the exim configuration are still
4790 caught, and the auth.log warnings about OTP are gone."
4792 PH/25 When debugging is enabled, the contents of the command line are added
4793 to the debugging output, even when log_selector=+arguments is not
4796 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4797 answer is "GNU", and only if the return is "GNU/something" is the answer
4800 PH/27 $acl_verify_message is now set immediately after the failure of a
4801 verification in an ACL, and so is available in subsequent modifiers. In
4802 particular, the message can be preserved by coding like this:
4804 warn !verify = sender
4805 set acl_m0 = $acl_verify_message
4807 Previously, $acl_verify_message was set only while expanding "message"
4808 and "log_message" when a very denied access.
4810 PH/28 Modified OS/os.c-Linux with
4812 -#ifndef OS_LOAD_AVERAGE
4813 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4815 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4816 nomenclature these days.)
4818 PH/29 Installed patch from the Sieve maintainer that adds the options
4819 sieve_useraddress and sieve_subaddress to the redirect router.
4821 PH/30 In these circumstances:
4822 . Two addresses routed to the same list of hosts;
4823 . First host does not offer TLS;
4824 . First host accepts first address;
4825 . First host gives temporary error to second address;
4826 . Second host offers TLS and a TLS session is established;
4827 . Second host accepts second address.
4828 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4829 and peerdn, if requested) that were in fact used only for the second
4832 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4833 attention to any local part prefix or suffix that was matched by the
4834 router that accepted the address. It now behaves in the same way as it
4835 does for delivery: the affixes are removed from the local part unless
4836 rcpt_include_affixes is set on the transport.
4838 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4839 timeout during the DATA phase of an incoming message.
4841 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4842 applied a patch sanctioned by the Sieve maintainer.
4844 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4845 the uid or gid is negative. A case of a negative gid caused this to be
4846 noticed. The fix allows for either to be negative.
4848 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4849 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4852 PH/36 The change PH/12 above was broken. Fixed it.
4854 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4855 the grounds that routing the same address twice would always produce the
4856 same answer. This might have been true once, but it is certainly no
4857 longer true now. Routing a child address may depend on the previous
4858 routing that produced that child. Some complicated redirection strategies
4859 went wrong when messages had multiple recipients, and made Exim's
4860 behaviour dependent on the order in which the addresses were given.
4862 I have moved the duplicate checking until after the routing is complete.
4863 Exim scans the addresses that are assigned to local and remote
4864 transports, and removes any duplicates. This means that more work will be
4865 done, as duplicates will always all be routed, but duplicates are
4866 presumably rare, so I don't expect this is of any significance.
4868 For deliveries to pipes, files, and autoreplies, the duplicate checking
4869 still happens during the routing process, since they are not going to be
4872 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4873 It corrects a timeout issue with spamd. This is Ian's comment: "The
4874 background is that sometimes spamd either never reads data from a
4875 connection it has accepted, or it never writes response data. The exiscan
4876 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4877 blindly assumes that writes won't block so it may never time out."
4879 PH/39 Allow G after quota size as well as K and M.
4881 PH/40 The value set for $authenticated_id in an authenticator may not contain
4882 binary zeroes or newlines because the value is written to log lines and
4883 to spool files. There was no check on this. Now the value is run through
4884 the string_printing() function so that such characters are converted to
4885 printable escape sequences.
4887 PH/41 $message_linecount is a new variable that contains the total number of
4888 lines in the message. Compare $body_linecount, which is the count for the
4891 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4892 bounce and delay warning messages, except in certain special cases, which
4895 (a) An SMTP error message from a remote host;
4896 (b) A message specified in a :fail: redirection;
4897 (c) A message specified in a "fail" command in a system filter;
4898 (d) A message specified in a FAIL return from the queryprogram router;
4899 (e) A message specified by the cannot_route_message router option.
4901 In these cases only, Exim does include the error details in bounce and
4902 warning messages. There are also a few cases where bland messages such
4903 as "unrouteable address" or "local delivery error" are given.
4905 PH/43 $value is now also set for the "else" part of a ${run expansion.
4907 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4908 being worked on, but at least Exim now implements the latest version to
4911 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4912 process to complete was treated as a delivery failure, a timeout while
4913 writing the message to the pipe was logged, but erroneously treated as a
4914 successful delivery. Such timeouts include transport filter timeouts. For
4915 consistency with the overall process timeout, these timeouts are now
4916 treated as errors, giving rise to delivery failures by default. However,
4917 there is now a new Boolean option for the pipe transport called
4918 timeout_defer, which, if set TRUE, converts the failures into defers for
4919 both kinds of timeout. A transport filter timeout is now identified in
4922 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4923 systems where "make" and "gmake" are different, calling "gmake" at top
4924 level broke things. I've arranged for the value of $(MAKE) to be passed
4925 from the Makefile to this script so that it can call the same version of
4929 A note about Exim versions 4.44 and 4.50
4930 ----------------------------------------
4932 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4933 changes of various kinds. As a consequence, a big documentation update was
4934 needed. This delayed the release for rather longer than seemed good, especially
4935 in the light of a couple of (minor) security issues. Therefore, the changes
4936 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4937 release. So 4.44 and 4.50 are in effect two different branches that both start
4940 I have left the 4.50 change log unchanged; it contains all the changes since
4941 4.43. The change log for 4.44 is below; many of its items are identical to
4942 those for 4.50. This seems to be the most sensible way to preserve the
4943 historical information.
4949 1. Minor wording change to the doc/README.SIEVE file.
4951 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4952 computation of the current number of files was incorrect.
4954 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4955 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4958 4. Give more explanation in the error message when the command for a transport
4959 filter fails to execute.
4961 5. There are several places where Exim runs a non-Exim command in a
4962 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4963 was being done only for the command run by the queryprogram router. It is
4964 now done for all such subprocesses. The other cases are: ${run, transport
4965 filters, and the commands run by the lmtp and pipe transports.
4967 6. Added CONFIGURE_GROUP build-time option.
4969 7. Some older OS have a limit of 256 on the maximum number of file
4970 descriptors. Exim was using setrlimit() to set 1000 as a large value
4971 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4972 systems. I've change it so that if it can't get 1000, it tries for 256.
4974 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4975 was an oversight, and furthermore, ever since the addition of extra
4976 controls (e.g. 4.43/32), the checks on when to allow different forms of
4977 "control" were broken. There should now be diagnostics for all cases when a
4978 control that does not make sense is encountered.
4980 9. Added the /retain_sender option to "control=submission".
4982 10. $recipients is now available in the predata ACL (oversight).
4984 11. Tidy the search cache before the fork to do a delivery from a message
4985 received from the command line. Otherwise the child will trigger a lookup
4986 failure and thereby defer the delivery if it tries to use (for example) a
4987 cached ldap connection that the parent has called unbind on.
4989 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4990 of $address_data from the recipient verification was clobbered by the
4991 sender verification.
4993 13. The value of address_data from a sender verification is now available in
4994 $sender_address_data in subsequent conditions in the ACL statement.
4996 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4998 15. Added a new option "connect=<time>" to callout options, to set a different
5001 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
5002 was its contents. (It was OK if the option was not defined at all.)
5004 17. A "Completed" log line is now written for messages that are removed from
5005 the spool by the -Mrm option.
5007 18. New variables $sender_verify_failure and $recipient_verify_failure contain
5008 information about exactly what failed.
5010 19. Added -dd to debug only the daemon process.
5012 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
5013 handles timeouts, both on the server side and network timeouts. Renamed the
5014 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
5016 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
5017 It is now set to "smtps".
5019 22. $host_address is now set to the target address during the checking of
5020 ignore_target_hosts.
5022 23. When checking ignore_target_hosts for an ipliteral router, no host name was
5023 being passed; this would have caused $sender_host_name to have been used if
5024 matching the list had actually called for a host name (not very likely,
5025 since this list is usually IP addresses). A host name is now passed as
5028 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5029 code that specifies a non-restarting handler (typically sigaction() in
5030 modern systems) in an attempt to fix a rare and obscure crash bug.
5032 25. Narrowed the window for a race in the daemon that could cause it to ignore
5033 SIGCHLD signals. This is not a major problem, because they are used only to
5034 wake it up if nothing else does.
5036 26. A malformed maildirsize file could cause Exim to calculate negative values
5037 for the mailbox size or file count. Odd effects could occur as a result.
5038 The maildirsize information is now recalculated if the size or filecount
5041 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5042 support for a long time. Removed HAVE_SYS_VFS_H.
5044 28. Installed the latest version of exipick from John Jetmore.
5046 29. In an address list, if the pattern was not a regular expression, an empty
5047 subject address (from a bounce message) matched only if the pattern was an
5048 empty string. Non-empty patterns were not even tested. This was the wrong
5049 because it is perfectly reasonable to use an empty address as part of a
5050 database query. An empty address is now tested by patterns that are
5051 lookups. However, all the other forms of pattern expect the subject to
5052 contain a local part and a domain, and therefore, for them, an empty
5053 address still always fails if the pattern is not itself empty.
5055 30. Exim went into a mad DNS loop when attempting to do a callout where the
5056 host was specified on an smtp transport, and looking it up yielded more
5057 than one IP address.
5059 31. Re-factored the code for checking spool and log partition space into a
5060 function that finds that data and another that does the check. The former
5061 is then used to implement four new variables: $spool_space, $log_space,
5062 $spool_inodes, and $log_inodes.
5064 32. The RFC2047 encoding function was originally intended for short strings
5065 such as real names; it was not keeping to the 75-character limit for
5066 encoded words that the RFC imposes. It now respects the limit, and
5067 generates multiple encoded words if necessary. To be on the safe side, I
5068 have increased the buffer size for the ${rfc2047: expansion operator from
5071 33. It is now permitted to omit both strings after an "if" condition; if the
5072 condition is true, the result is "true". As before, when the second string
5073 is omitted, a false condition yields an empty string. This makes it less
5074 cumbersome to write custom ACL and router conditions.
5076 34. Failure to deliver a bounce message always caused it to be frozen, even if
5077 there was an errors_to setting on the router. The errors_to setting is now
5080 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5081 canonical form (fully expanded) before being placed in
5082 $sender_host_address.
5084 36. The table in the code that translates DNS record types into text (T_A to
5085 "A" for instance) was missing entries for NS and CNAME. It is just possible
5086 that this could have caused confusion if both these types were looked up
5087 for the same domain, because the text type is used as part of Exim's
5088 per-process caching. But the chance of anyone hitting this buglet seems
5091 37. The dnsdb lookup has been extended in a number of ways.
5093 (1) There is a new type, "zns", which walks up the domain tree until it
5094 finds some nameserver records. It should be used with care.
5096 (2) There is a new type, "mxh", which is like "mx" except that it returns
5097 just the host names, not the priorities.
5099 (3) It is now possible to give a list of domains (or IP addresses) to be
5100 looked up. The behaviour when one of the lookups defers can be
5101 controlled by a keyword.
5103 (4) It is now possible to specify the separator character for use when
5104 multiple records are returned.
5106 38. The dnslists ACL condition has been extended: it is now possible to supply
5107 a list of IP addresses and/or domains to be looked up in a particular DNS
5110 39. Added log_selector=+queue_time_overall.
5112 40. When running the queue in the test harness, wait just a tad after forking a
5113 delivery process, to get repeatability of debugging output.
5115 41. Include certificate and key file names in error message when GnuTLS fails
5116 to set them up, because the GnuTLS error message doesn't include the name
5117 of the failing file when there is a problem reading it.
5119 42. Allow both -bf and -bF in the same test run.
5121 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
5123 44. The "Exiscan patch" is now merged into the mainline Exim source.
5125 45. Sometimes the final signoff response after QUIT could fail to get
5126 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5127 < 0 before doing a fflush(). This bug looks as though it goes back to the
5128 introduction of TLS in release 3.20, but "sometimes" must have been rare
5129 because the tests only now provoked it.
5131 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
5132 (this can affect the format of dates).
5134 47. exim_tidydb, when checking for the continued existence of a message for
5135 which it has found a message-specific retry record, was not finding
5136 messages that were in split spool directories. Consequently, it was
5137 deleting retry records that should have stayed in existence.
5139 48. Steve fixed some bugs in eximstats.
5141 49. The SPA authentication driver was not abandoning authentication and moving
5142 on to the next authenticator when an expansion was forced to fail,
5143 contradicting the general specification for all authenticators. Instead it
5144 was generating a temporary error. It now behaves as specified.
5146 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
5147 (the order specifies the preference for clients). The order is now AES256,
5148 AES128, 3DES, ARCFOUR128.
5150 51. Small patch to Sieve code - explicitly set From: when generating an
5153 52. Exim crashed if a remote delivery caused a very long error message to be
5154 recorded - for instance if somebody sent an entire SpamAssassin report back
5155 as a large number of 550 error lines. This bug was coincidentally fixed by
5156 increasing the size of one of Exim's internal buffers (big_buffer) that
5157 happened as part of the Exiscan merge. However, to be on the safe side, I
5158 have made the code more robust (and fixed the comments that describe what
5161 53. Now that there can be additional text after "Completed" in log lines (if
5162 the queue_time_overall log selector is set), a one-byte patch to exigrep
5163 was needed to allow it to recognize "Completed" as not the last thing in
5166 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
5167 patch that reportedly fixes this has been added. I am not expert enough to
5168 create a test for it. This is what the patch creator wrote:
5170 "I found a little strange behaviour of ldap code when working with
5171 Windows 2003 AD Domain, where users was placed in more than one
5172 Organization Units. When I tried to give exim partial DN, the exit code
5173 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
5174 But simultaneously result of request was absolutely normal ldap result,
5175 so I produce this patch..."
5177 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
5178 REFERENCE, so I have modified the code to exclude the patch when that macro
5181 55. Some experimental protocols are using DNS PTR records for new purposes. The
5182 keys for these records are domain names, not reversed IP addresses. The
5183 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5184 leaves it alone. Component reversal etc. now happens only for IP addresses.
5187 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
5189 57. Double the size of the debug message buffer (to 2048) so that more of very
5190 long debug lines gets shown.
5192 58. The exicyclog utility now does better if the number of log files to keep
5193 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
5195 59. Two changes related to the smtp_active_hostname option:
5197 (1) $smtp_active_hostname is now available as a variable.
5198 (2) The default for smtp_banner uses $smtp_active_hostname instead
5199 of $primary_hostname.
5201 60. The host_aton() function is supposed to be passed a string that is known
5202 to be a valid IP address. However, in the case of IPv6 addresses, it was
5203 not checking this. This is a hostage to fortune. Exim now panics and dies
5204 if the condition is not met. A case was found where this could be provoked
5205 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5206 components; fortuitously, this particular loophole had already been fixed
5207 by change 4.50/55 above.
5209 If there are any other similar loopholes, the new check in host_aton()
5210 itself should stop them being exploited. The report I received stated that
5211 data on the command line could provoke the exploit when Exim was running as
5212 exim, but did not say which command line option was involved. All I could
5213 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5214 running as the user.
5217 61. There was a buffer overflow vulnerability in the SPA authentication code
5218 (which came originally from the Samba project). I have added a test to the
5219 spa_base64_to_bits() function which I hope fixes it.
5222 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
5223 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
5225 63. The daemon start-up calls getloadavg() while still root for those OS that
5226 need the first call to be done as root, but it missed one case: when
5227 deliver_queue_load_max is set with deliver_drop_privilege. This is
5228 necessary for the benefit of the queue runner, because there is no re-exec
5229 when deliver_drop_privilege is set.
5231 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
5232 This has been fixed.
5234 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5235 was in use, was not putting the data itself into the right store pool;
5236 consequently, it could be overwritten for a subsequent message in the same
5237 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5240 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
5242 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
5243 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
5244 the function were treating the return as a boolean value, which happened to
5245 work because 0=false and not-0=true, but is not correct code.
5247 68. The host_aton() function was not handling scoped IPv6 addresses (those
5248 with, for example, "%eth0" on the end) correctly.
5250 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
5251 negated items (that is, ~something) in unsigned ints. Some compilers
5252 apparently mutter when there is no cast.
5254 70. If an address verification called from an ACL failed, and did not produce a
5255 user-specific message (i.e. there was only a "system" message), nothing was
5256 put in $acl_verify_message. In this situation, it now puts the system
5259 71. Change 4.23/11 added synchronization checking at the start of an SMTP
5260 session; change 4.31/43 added the unwanted input to the log line - except
5261 that it did not do this in the start of session case. It now does.
5263 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
5264 This is wrong and can cause the other end to generate a synchronization
5265 error if it is another Exim or anything else that does the synchronization
5266 check. A QUIT command is no longer sent after a timeout.
5268 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
5269 during host lookups.
5271 74. The defer_ok option of callout verification was not working if it was used
5272 when verifying addresses in header lines, that is, for this case:
5274 verify = header_sender/callout=defer_ok
5276 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
5277 those file descriptors could be used for SMTP connections. If anything
5278 wrote to stderr (the example that came up was "warn" in embedded Perl), it
5279 could be sent to the SMTP client, causing chaos. The daemon now opens
5280 stdin, stdout, and stderr to /dev/null when it puts itself into the
5283 76. Arrange for output from Perl's "warn" command to be written to Exim's main
5284 log by default. The user can override this with suitable Perl magic.
5286 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
5287 the log message when discard triggers, was not working for the DATA ACL or
5288 for the non-SMTP ACL.
5290 78. Error message wording change in sieve.c.
5292 79. If smtp_accept_max_per_host was set, the number of connections could be
5293 restricted to fewer than expected, because the daemon was trying to set up
5294 a new connection before checking whether the processes handling previous
5295 connections had finished. The check for completed processes is now done
5296 earlier. On busy systems, this bug wouldn't be noticed because something
5297 else would have woken the daemon, and it would have reaped the completed
5300 80. If a message was submitted locally by a user whose login name contained one
5301 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
5302 It caused a spool format error. I have fixed the spool reading code. A
5303 related problem was that the "from" clause in the Received: line became
5304 illegal because of the space(s). It is now covered by ${quote_local_part.
5306 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
5309 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
5311 83. Patch from Sieve maintainer to fix unterminated string problem in
5312 "vacation" handling.
5314 84. Some minor changes to the Linux configuration files to help with other
5315 OS variants using glibc.
5317 85. One more patch for Sieve to update vacation handling to latest spec.
5320 ----------------------------------------------------
5321 See the note above about the 4.44 and 4.50 releases.
5322 ----------------------------------------------------
5328 1. Change 4.43/35 introduced a bug that caused file counts to be
5329 incorrectly computed when quota_filecount was set in an appendfile
5332 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
5333 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
5336 3. Give more explanation in the error message when the command for a transport
5337 filter fails to execute.
5339 4. There are several places where Exim runs a non-Exim command in a
5340 subprocess. The SIGUSR1 signal should be disabled for these processes. This
5341 was being done only for the command run by the queryprogram router. It is
5342 now done for all such subprocesses. The other cases are: ${run, transport
5343 filters, and the commands run by the lmtp and pipe transports.
5345 5. Some older OS have a limit of 256 on the maximum number of file
5346 descriptors. Exim was using setrlimit() to set 1000 as a large value
5347 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
5348 systems. I've change it so that if it can't get 1000, it tries for 256.
5350 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
5351 was an oversight, and furthermore, ever since the addition of extra
5352 controls (e.g. 4.43/32), the checks on when to allow different forms of
5353 "control" were broken. There should now be diagnostics for all cases when a
5354 control that does not make sense is encountered.
5356 7. $recipients is now available in the predata ACL (oversight).
5358 8. Tidy the search cache before the fork to do a delivery from a message
5359 received from the command line. Otherwise the child will trigger a lookup
5360 failure and thereby defer the delivery if it tries to use (for example) a
5361 cached ldap connection that the parent has called unbind on.
5363 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
5364 of $address_data from the recipient verification was clobbered by the
5365 sender verification.
5367 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
5368 was its contents. (It was OK if the option was not defined at all.)
5370 11. A "Completed" log line is now written for messages that are removed from
5371 the spool by the -Mrm option.
5373 12. $host_address is now set to the target address during the checking of
5374 ignore_target_hosts.
5376 13. When checking ignore_target_hosts for an ipliteral router, no host name was
5377 being passed; this would have caused $sender_host_name to have been used if
5378 matching the list had actually called for a host name (not very likely,
5379 since this list is usually IP addresses). A host name is now passed as
5382 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5383 code that specifies a non-restarting handler (typically sigaction() in
5384 modern systems) in an attempt to fix a rare and obscure crash bug.
5386 15. Narrowed the window for a race in the daemon that could cause it to ignore
5387 SIGCHLD signals. This is not a major problem, because they are used only to
5388 wake it up if nothing else does.
5390 16. A malformed maildirsize file could cause Exim to calculate negative values
5391 for the mailbox size or file count. Odd effects could occur as a result.
5392 The maildirsize information is now recalculated if the size or filecount
5395 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5396 support for a long time. Removed HAVE_SYS_VFS_H.
5398 18. Updated exipick to current release from John Jetmore.
5400 19. Allow an empty sender to be matched against a lookup in an address list.
5401 Previously the only cases considered were a regular expression, or an
5404 20. Exim went into a mad DNS lookup loop when doing a callout where the
5405 host was specified on the transport, if the DNS lookup yielded more than
5408 21. The RFC2047 encoding function was originally intended for short strings
5409 such as real names; it was not keeping to the 75-character limit for
5410 encoded words that the RFC imposes. It now respects the limit, and
5411 generates multiple encoded words if necessary. To be on the safe side, I
5412 have increased the buffer size for the ${rfc2047: expansion operator from
5415 22. Failure to deliver a bounce message always caused it to be frozen, even if
5416 there was an errors_to setting on the router. The errors_to setting is now
5419 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5420 canonical form (fully expanded) before being placed in
5421 $sender_host_address.
5423 24. Updated eximstats to version 1.33
5425 25. Include certificate and key file names in error message when GnuTLS fails
5426 to set them up, because the GnuTLS error message doesn't include the name
5427 of the failing file when there is a problem reading it.
5429 26. Expand error message when OpenSSL has problems setting up cert/key files.
5432 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
5433 (this can affect the format of dates).
5435 28. exim_tidydb, when checking for the continued existence of a message for
5436 which it has found a message-specific retry record, was not finding
5437 messages that were in split spool directories. Consequently, it was
5438 deleting retry records that should have stayed in existence.
5440 29. eximstats updated to version 1.35
5441 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
5442 1.35 - bugfix such that pie charts by volume are generated correctly
5444 30. The SPA authentication driver was not abandoning authentication and moving
5445 on to the next authenticator when an expansion was forced to fail,
5446 contradicting the general specification for all authenticators. Instead it
5447 was generating a temporary error. It now behaves as specified.
5449 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
5450 (the order specifies the preference for clients). The order is now AES256,
5451 AES128, 3DES, ARCFOUR128.
5453 31. Small patch to Sieve code - explicitly set From: when generating an
5456 32. Exim crashed if a remote delivery caused a very long error message to be
5457 recorded - for instance if somebody sent an entire SpamAssassin report back
5458 as a large number of 550 error lines. This bug was coincidentally fixed by
5459 increasing the size of one of Exim's internal buffers (big_buffer) that
5460 happened as part of the Exiscan merge. However, to be on the safe side, I
5461 have made the code more robust (and fixed the comments that describe what
5464 33. Some experimental protocols are using DNS PTR records for new purposes. The
5465 keys for these records are domain names, not reversed IP addresses. The
5466 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5467 leaves it alone. Component reversal etc. now happens only for IP addresses.
5470 34. The host_aton() function is supposed to be passed a string that is known
5471 to be a valid IP address. However, in the case of IPv6 addresses, it was
5472 not checking this. This is a hostage to fortune. Exim now panics and dies
5473 if the condition is not met. A case was found where this could be provoked
5474 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5475 components; fortuitously, this particular loophole had already been fixed
5476 by change 4.50/55 or 4.44/33 above.
5478 If there are any other similar loopholes, the new check in host_aton()
5479 itself should stop them being exploited. The report I received stated that
5480 data on the command line could provoke the exploit when Exim was running as
5481 exim, but did not say which command line option was involved. All I could
5482 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5483 running as the user.
5486 35. There was a buffer overflow vulnerability in the SPA authentication code
5487 (which came originally from the Samba project). I have added a test to the
5488 spa_base64_to_bits() function which I hope fixes it.
5491 36. The daemon start-up calls getloadavg() while still root for those OS that
5492 need the first call to be done as root, but it missed one case: when
5493 deliver_queue_load_max is set with deliver_drop_privilege. This is
5494 necessary for the benefit of the queue runner, because there is no re-exec
5495 when deliver_drop_privilege is set.
5497 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5498 was in use, was not putting the data itself into the right store pool;
5499 consequently, it could be overwritten for a subsequent message in the same
5500 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5503 38. Sometimes the final signoff response after QUIT could fail to get
5504 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5505 < 0 before doing a fflush(). This bug looks as though it goes back to the
5506 introduction of TLS in release 3.20, but "sometimes" must have been rare
5507 because the tests only now provoked it.
5513 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5514 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5515 Now it returns an integer. A number of calls were still expecting a T/F
5516 return. Fortuitously, in all cases, the tests worked in OK situations,
5517 which is the norm. However, things would have gone wrong on any write
5518 failures on the smtp file descriptor. This function is used when sending
5519 messages over SMTP and also when doing verify callouts.
5521 2. When Exim is called to do synchronous delivery of a locally submitted
5522 message (the -odf or -odi options), it no longer closes stderr before doing
5525 3. Implemented the mua_wrapper option.
5527 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5529 5. Implemented the functions header_remove(), header_testname(),
5530 header_add_at_position(), and receive_remove_recipient(), and exported them
5533 6. If an ACL "warn" statement specified the addition of headers, Exim already
5534 inserted X-ACL-Warn: at the start if there was no header name. However, it
5535 was not making this test for the second and subsequent header lines if
5536 there were newlines in the string. This meant that an invalid header could
5537 be inserted if Exim was badly configured.
5539 7. Allow an ACL "warn" statement to add header lines at the start or after all
5540 the Received: headers, as well as at the end.
5542 8. Added the rcpt_4xx retry error code.
5544 9. Added postmaster_mailfrom=xxx to callout verification option.
5546 10. Added mailfrom=xxxx to the callout verification option, for verify=
5549 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5550 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5552 12. Inserted some casts to stop certain compilers warning when using pointer
5553 differences as field lengths or precisions in printf-type calls (mostly
5554 affecting debugging statements).
5556 13. Added optional readline() support for -be (dynamically loaded).
5558 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5559 same clock tick as a message's arrival, so that its received time was the
5560 same as the "first fail" time on the retry record, and that message
5561 remained on the queue past the ultimate address timeout, every queue runner
5562 would try a delivery (because it was past the ultimate address timeout) but
5563 after another failure, the ultimate address timeout, which should have then
5564 bounced the address, did not kick in. This was a "< instead of <=" error;
5565 in most cases the first failure would have been in the next clock tick
5566 after the received time, and all would be well.
5568 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5569 being recognized when the domain list was tested by the match_domain
5570 condition in an expansion string.
5572 16. Added the ${str2b64: operator.
5574 17. Exim was always calling setrlimit() to set a large limit for the number of
5575 processes, without checking whether the existing limit was already
5576 adequate. (It did check for the limit on file descriptors.) Furthermore,
5577 errors from getrlimit() and setrlimit() were being ignored. Now they are
5578 logged to the main and panic logs, but Exim does carry on, to try to do its
5579 job under whatever limits there are.
5581 18. Imported PCRE 5.0.
5583 19. Trivial typo in log message " temporarily refused connection" (the leading
5586 20. If the log selector return_path_on_delivery was set and an address was
5587 redirected to /dev/null, the delivery process crashed because it assumed
5588 that a return path would always be set for a "successful" delivery. In this
5589 case, the whole delivery is bypassed as an optimization, and therefore no
5592 21. Internal re-arrangement: the function for sending a challenge and reading
5593 a response while authentication was assuming a zero-terminated challenge
5594 string. It's now changed to take a pointer and a length, to allow for
5595 binary data in such strings.
5597 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5599 23. Exim was not respecting finduser_retries when seeking the login of the
5600 uid under which it was called; it was always trying 10 times. (The default
5601 setting of finduser_retries is zero.) Also, it was sleeping after the final
5602 failure, which is pointless.
5604 24. Implemented tls_on_connect_ports.
5606 25. Implemented acl_smtp_predata.
5608 26. If the domain in control=submission is set empty, Exim assumes that the
5609 authenticated id is a complete email address when it generates From: or
5610 Sender: header lines.
5612 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5613 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5614 chown and chgrp in /bin and hostname in /usr/bin.
5616 28. Exim was keeping the "process log" file open after each use, just as it
5617 does for the main log. This opens the possibility of it remaining open for
5618 long periods when the USR1 signal hits a daemon. Occasional processlog
5619 errors were reported, that could have been caused by this. Anyway, it seems
5620 much more sensible not to leave this file open at all, so that is what now
5623 29. The long-running daemon process does not normally write to the log once it
5624 has entered its main loop, and it closes the log before doing so. This is
5625 so that log files can straightforwardly be renamed and moved. However,
5626 there are a couple of unusual error situations where the daemon does write
5627 log entries, and I had neglected to close the log afterwards.
5629 30. The text of an SMTP error response that was received during a remote
5630 delivery was being truncated at 512 bytes. This is too short for some of
5631 the long messages that one sometimes sees. I've increased the limit to
5634 31. It is now possible to make retry rules that apply only when a message has a
5635 specific sender, in particular, an empty sender.
5637 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5638 it possible to be selective about when SMTP synchronization is enforced.
5640 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5642 32. Implemented hosts_connection_nolog.
5644 33. Added an ACL for QUIT.
5646 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5649 35. Added mailbox_size and mailbox_filecount to appendfile.
5651 36. Added control = no_multiline_responses to ACLs.
5653 37. There was a bug in the logic of the code that waits for the clock to tick
5654 in the case where the clock went backwards by a substantial amount such
5655 that the microsecond fraction of "now" was more than the microsecond
5656 fraction of "then" (but the whole seconds number was less).
5658 38. Added support for the libradius Radius client library this is found on
5659 FreeBSD (previously only the radiusclient library was supported).
5665 1. When certain lookups returned multiple values in the form name=value, the
5666 quoting of the values was not always being done properly. Specifically:
5667 (a) If the value started with a double quote, but contained no whitespace,
5669 (b) If the value contained whitespace other than a space character (i.e.
5670 tabs or newlines or carriage returns) it was not quoted.
5671 This fix has been applied to the mysql and pgsql lookups by writing a
5672 separate quoting function and calling it from the lookup code. The fix
5673 should probably also be applied to nisplus, ibase and oracle lookups, but
5674 since I cannot test any of those, I have not disturbed their existing code.
5676 2. A hit in the callout cache for a specific address caused a log line with no
5677 reason for rejecting RCPT. Now it says "Previous (cached) callout
5678 verification failure".
5680 3. There was an off-by-one bug in the queryprogram router. An over-long
5681 return line was truncated at 256 instead of 255 characters, thereby
5682 overflowing its buffer with the terminating zero. As well as fixing this, I
5683 have increased the buffer size to 1024 (and made a note to document this).
5685 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5686 when Exim is waiting for an SMTP response from a remote server, Exim
5687 restarts its select() call on the socket, thereby resetting its timeout.
5688 This is not a problem when such interrupts are rare. Somebody set up a cron
5689 job to run exiwhat every 2 minutes, which is less than the normal select()
5690 timeout (5 or 10 minutes). This meant that the select() timeout never
5691 kicked in because it was always reset. I have fixed this by comparing the
5692 time when an interrupt arrives with the time at the start of the first call
5693 to select(). If more time than the timeout has elapsed, the interrupt is
5694 treated as a timeout.
5696 5. Some internal re-factoring in preparation for the addition of Sieve
5697 extensions (by MH). In particular, the "personal" test is moved to a
5698 separate function, and given an option for scanning Cc: and Bcc: (which is
5699 not set for Exim filters).
5701 6. When Exim created an email address using the login of the caller as the
5702 local part (e.g. when creating a From: or Sender: header line), it was not
5703 quoting the local part when it contained special characters such as @.
5705 7. Installed new OpenBSD configuration files.
5707 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5708 try to make them clearer.
5710 9. Callout options, other than the timeout value, were being ignored when
5711 verifying sender addresses in header lines. For example, when using
5713 verify = header_sender/callout=no_cache
5715 the cache was (incorrectly) being used.
5717 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5718 only the Cygwin environment.
5720 11. When return_path_on_delivery was set as a log selector, if different remote
5721 addresses in the same message used different return paths and parallel
5722 remote delivery occurred, the wrong values would sometimes be logged.
5723 (Whenever a remote delivery process finished, the return path value from
5724 the most recently started remote delivery process was logged.)
5726 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5727 lines when AUTH and/or TLS are in use. This is the "received protocol"
5728 field. Exim used to use "asmtp" for authenticated SMTP, without any
5729 indication (in the protocol name) for TLS use. Now it follows the RFC and
5730 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5731 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5732 names appear in log lines as well as in Received: header lines.
5734 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5735 extensions, and comparison tests, and to fix some bugs.
5737 14. Changes to the "personal" filter test:
5739 (1) The test was buggy in that it was just doing the equivalent of
5740 "contains" tests on header lines. For example, if a user's address was
5741 anne@some.where, the "personal" test would incorrectly be true for
5743 To: susanne@some.where
5745 This test is now done by extracting each address from the header in turn,
5746 and checking the entire address. Other tests that are part of "personal"
5747 are now done using regular expressions (for example, to check local parts
5748 of addresses in From: header lines).
5750 (2) The list of non-personal local parts in From: addresses has been
5751 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5752 taken from the Sieve specification recommendations.
5754 (3) If the message contains any header line starting with "List-" it is
5755 treated as non-personal.
5757 (4) The test for "circular" in the Subject: header line has been removed
5758 because it now seems ill-conceived.
5760 15. Minor typos in src/EDITME comments corrected.
5762 16. Installed latest exipick from John Jetmore.
5764 17. If headers_add on a router specified a text string that was too long for
5765 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5766 of string_sprintf() is now avoided.
5768 18. $message_body_size was not set (it was always zero) when running the DATA
5769 ACL and the local_scan() function.
5771 19. For the "mail" command in an Exim filter, no default was being set for
5772 the once_repeat time, causing a random time value to be used if "once" was
5773 specified. (If the value happened to be <= 0, no repeat happened.) The
5774 default is now 0s, meaning "never repeat". The "vacation" command was OK
5775 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5776 (I found it when inspecting the code).
5778 20. There is now an overall timeout for performing a callout verification. It
5779 defaults to 4 times the callout timeout, which applies to individual SMTP
5780 commands during the callout. The overall timeout applies when there is more
5781 than one host that can be tried. The timeout is checked before trying the
5782 next host. This prevents very long delays if there are a large number of
5783 hosts and all are timing out (e.g. when the network connections are timing
5784 out). The value of the overall timeout can be changed by specifying an
5785 additional sub-option for "callout", called "maxwait". For example:
5787 verify = sender/callout=5s,maxwait=20s
5789 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5790 to the end before writing, but this should make it even safer).
5792 22. Exim was forgetting that it had advertised PIPELINING for the second and
5793 subsequent messages on an SMTP connection. It was also not resetting its
5794 memory on STARTTLS and an internal HELO.
5796 23. When Exim logs an SMTP synchronization error within a session, it now
5797 records whether PIPELINING has been advertised or not.
5799 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5800 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5801 rather than long int.
5803 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5805 26. Added the never_mail option to autoreply.
5811 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5812 crash if the getsockname() call failed; this can happen if a connection is
5813 closed very soon after it is established. The problem was simply in the
5814 order in which certain operations were done, causing Exim to try to write
5815 to the SMTP stream before it had set up the file descriptor. The bug has
5816 been fixed by making things happen in the correct order.
5822 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5823 before the connection was closed, thus losing the rejection response.
5825 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5826 some early Solaris releases, but causes trouble in current releases where
5827 socklen_t is defined.
5829 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5832 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5835 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5836 that releasing the top of it at the end releases what was used for sub-
5837 expansions (unless the block got too big). However, discard this block if
5838 the first thing is a variable or header, so that we can use its block when
5839 it is dynamic (useful for very large $message_headers, for example).
5841 6. Lookups now cache *every* query, not just the most recent. A new, separate
5842 store pool is used for this. It can be recovered when all lookup caches are
5843 flushed. Lookups now release memory at the end of their result strings.
5844 This has involved some general refactoring of the lookup sources.
5846 7. Some code has been added to the store_xxx() functions to reduce the amount
5847 of flapping under certain conditions.
5849 8. log_incoming_interface used to affect only the <= reception log lines. Now
5850 it causes the local interface and port to be added to several more SMTP log
5851 lines, for example "SMTP connection from", and rejection lines.
5853 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5855 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5857 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5858 could be overwritten at the end of the current message (or the start of a
5859 new message if it was set in a HELO ACL). The value is now preserved for
5860 the duration of the SMTP connection.
5862 12. If a transport had a headers_rewrite setting, and a matching header line
5863 contained an unqualified address, that address was qualified, even if it
5864 did not match any rewriting rules. The underlying bug was that the values
5865 of the flags that permit the existence of unqualified sender and recipient
5866 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5867 non-local messages, and by -bnq for local messages) were not being
5868 preserved with the message after it was received.
5870 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5871 "next input=" as part of the text comprising the host identity instead of
5872 the correct text. The code was using the same buffer for two different
5873 strings. However, depending on which order the printing function evaluated
5874 its arguments, the bug did not always show up. Under Linux, for example, my
5875 test suite worked just fine.
5877 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5878 doesn't work with some older versions of Perl. It has been changed to "my",
5879 which in any case is probably the better facility to use.
5881 15. A really picky compiler found some instances of statements for creating
5882 error messages that either had too many or two few arguments for the format
5885 16. The size of the buffer for calls to the DNS resolver has been increased
5886 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5887 for addresses that have a lot of PTR records. This alleviates a problem; it
5888 does not fully solve it.
5890 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5891 buffer now truncates the list and logs the incident, which is the same
5892 action as happens when Exim is looking up a host name and its aliases.
5893 Previously in this situation something unpredictable would happen;
5894 sometimes it was "internal error: store_reset failed".
5896 18. If a server dropped the connection unexpectedly when an Exim client was
5897 using GnuTLS and trying to read a response, the client delivery process
5898 crashed while trying to generate an error log message.
5900 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5901 string, for example:
5903 warn message = H1: something\nH2: something
5905 the text was added as a single header line from Exim's point of view
5906 though it ended up OK in the delivered message. However, searching for the
5907 second and subsequent header lines using $h_h2: did not work. This has been
5908 fixed. Similarly, if a system filter added multiple headers in this way,
5909 the routers could not see them.
5911 20. Expanded the error message when iplsearch is called with an invalid key to
5912 suggest using net-iplsearch in a host list.
5914 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5915 ACLs are no longer actually imposed (and a message to that effect is
5918 22. If a "gecos" field in a passwd entry contained escaped characters, in
5919 particular, if it contained a \" sequence, Exim got it wrong when building
5920 a From: or a Sender: header from that name. A second bug also caused
5921 incorrect handling when an unquoted " was present following a character
5922 that needed quoting.
5924 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5925 was not being matched caselessly.
5927 24. Arranged for all hyphens in the exim.8 source to be escaped with
5930 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5931 properly. Recipient callout cache records were still being keyed to include
5932 the sender, even when use_sender was set false. This led to far more
5933 callouts that were necessary. The sender is no longer included in the key
5934 when use_sender is false.
5936 26. Added "control = submission" modifier to ACLs.
5938 27. Added the ${base62d: operator to decode base 62 numbers.
5940 28. dnsdb lookups can now access SRV records.
5942 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5943 the configuration file.
5945 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5946 (-v) mode. This makes the output for a verbose queue run more intelligible.
5948 31. Added a use_postmaster feature to recipient callouts.
5950 32. Added the $body_zerocount variable, containing the number of binary zero
5951 bytes in the message body.
5953 33. The time of last modification of the "new" subdirectory is now used as the
5954 "mailbox time last read" when there is a quota error for a maildir
5957 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5959 35. Added +ignore_unknown as a special item in host lists.
5961 36. Code for decoding IPv6 addresses in host lists is now included, even if
5962 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5963 address was recognized as an IP address, but was then not correctly decoded
5964 into binary, causing unexpected and incorrect effects when compared with
5971 1. Very minor rewording of debugging text in manualroute to say "list of
5972 hosts" instead of "hostlist".
5974 2. If verify=header_syntax was set, and a header line with an unqualified
5975 address (no domain) and a large number of spaces between the end of the
5976 name and the colon was received, the reception process suffered a buffer
5977 overflow, and (when I tested it) crashed. This was caused by some obsolete
5978 code that should have been removed. The fix is to remove it!
5980 3. When running in the test harness, delay a bit after writing a bounce
5981 message to get a bit more predictability in the log output.
5983 4. Added a call to search_tidyup() just before forking a reception process. In
5984 theory, someone could use a lookup in the expansion of smtp_accept_max_
5985 per_host which, without the tidyup, could leave open a database connection.
5987 5. Added the variables $recipient_data and $sender_data which get set from a
5988 lookup success in an ACL "recipients" or "senders" condition, or a router
5989 "senders" option, similar to $domain_data and $local_part_data.
5991 6. Moved the writing of debug_print from before to after the "senders" test
5994 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5995 problems for message scanning, either using a data ACL, or using
5996 local_scan() because the Received: header was not generated till after they
5997 were called (in order to set the time as the time of reception completion).
5998 I have revised the way this works. The header is now generated after the
5999 body is received, but before the ACL or local_scan() are called. After they
6000 are run, the timestamp in the header is updated.
6006 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
6007 before starting a queue runner without re-exec. This happened only when
6008 deliver_drop_privilege was set or when the Exim user was set to root. The
6009 effect of the bug was that timeouts during subsequent deliveries caused
6010 crashes instead of being properly handled. The handler is now left at its
6011 default (and expected) setting.
6013 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
6014 message, again when deliver_drop_privilege is set or Exim is run as root.
6015 The bug described in (1) was not present in this case, but the tidying up
6016 of the other signals was missing. I have made the two cases consistent.
6018 3. The ignore_target_hosts setting on a manualroute router was being ignored
6019 for hosts that were looked up using the /MX notation.
6021 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
6024 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
6025 operated on the sender address. After changing the $sender_address to <>
6026 for the sender address verify, Exim was re-instated it as the original
6027 (before rewriting) address, but remembering that it had rewritten it, so it
6028 wasn't rewriting it again. This bug also had the effect of breaking the
6029 sender address verification caching when the sender address was rewritten.
6031 6. The ignore_target_hosts option was being ignored by the ipliteral router.
6032 This has been changed so that if the ip literal address matches
6033 ignore_target_hosts, the router declines.
6035 7. Added expansion conditions match_domain, match_address, and match_local_
6036 part (NOT match_host).
6038 8. The placeholder for the Received: header didn't have a length field set.
6040 9. Added code to Exim itself and to exim_lock to test for a specific race
6041 condition that could lead to file corruption when using MBX delivery. The
6042 issue is with the lockfile that is created in /tmp. If this file is removed
6043 after a process has opened it but before that process has acquired a lock,
6044 there is the potential for a second process to recreate the file and also
6045 acquire a lock. This could lead to two Exim processes writing to the file
6046 at the same time. The added code performs the same test as UW imapd; it
6047 checks after acquiring the lock that its file descriptor still refers to
6048 the same named file.
6050 10. The buffer for building added header lines was of fixed size, 8192 bytes.
6051 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
6054 11. Added the smtp_active_hostname option. If used, this will typically be made
6055 to depend on the incoming interface address. Because $interface_address is
6056 not set up until the daemon has forked a reception process, error responses
6057 that can happen earlier (such as "too many connections") no longer contain
6060 12. If an expansion in a condition on a "warn" statement fails because a lookup
6061 defers, the "warn" statement is abandoned, and the next ACL statement is
6062 processed. Previously this caused the whole ACL to be aborted.
6064 13. Added the iplsearch lookup type.
6066 14. Added ident_timeout as a log selector.
6068 15. Added tls_certificate_verified as a log selector.
6070 16. Added a global option tls_require_ciphers (compare the smtp transport
6071 option of the same name). This controls incoming TLS connections.
6073 17. I finally figured out how to make tls_require_ciphers do a similar thing
6074 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
6075 before starting the TLS session.
6077 18. Tabs are now shown as \t in -bP output.
6079 19. If the log selector return_path_on_delivery was set, Exim crashed when
6080 bouncing a message because it had too many Received: header lines.
6082 20. If two routers both had headers_remove settings, and the first one included
6083 a superfluous trailing colon, the final name in the first list and the
6084 first name in the second list were incorrectly joined into one item (with a
6085 colon in the middle).
6091 1. Added -C and -D options to the exinext utility, mainly to make it easier
6092 to include in the automated testing, but these could be helpful when
6093 multiple configurations are in use.
6095 2. The exinext utility was not formatting the output nicely when there was
6096 an alternate port involved in the retry record key, nor when there was a
6097 message id as well (for retries that were specific to a specific message
6098 and a specific host). It was also confused by IPv6 addresses, because of
6099 the additional colons they contain. I have fixed the IPv4 problem, and
6100 patched it up to do a reasonable job for IPv6.
6102 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
6103 delivery, the log line now contains "pipelined" if PIPELINING was used.
6105 4. An SMTP transport process used to panic and die if the bind() call to set
6106 an explicit outgoing interface failed. This has been changed; it is now
6107 treated in the same way as a connect() failure.
6109 5. A reference to $sender_host_name in the part of a conditional expansion
6110 that was being skipped was still causing a DNS lookup. This no longer
6113 6. The def: expansion condition was not recognizing references to header lines
6114 that used bh_ and bheader_.
6116 7. Added the _cache feature to named lists.
6118 8. The code for checking quota_filecount in the appendfile transport was
6119 allowing one more file than it should have been.
6121 9. For compatibility with Sendmail, the command line option
6129 and sets the incoming protocol and host name (for trusted callers). The
6130 host name and its colon can be omitted when only the protocol is to be set.
6131 Note the Exim already has two private options, -pd and -ps, that refer to
6132 embedded Perl. It is therefore impossible to set a protocol value of "d" or
6133 "s", but I don't think that's a major issue.
6135 10. A number of refactoring changes to the code, none of which should affect
6138 (a) The number of logging options was getting close to filling up the
6139 32-bit word that was used as a bit map. I have split them into two classes:
6140 those that are passed in the argument to log_write(), and those that are
6141 only ever tested independently outside of that function. These are now in
6142 separate 32-bit words, so there is plenty of room for expansion again.
6143 There is no change in the user interface or the logging behaviour.
6145 (b) When building, for example, log lines, the code previously used a
6146 macro that called string_cat() twice, in order to add two strings. This is
6147 not really sufficiently general. Furthermore, there was one instance where
6148 it was actually wrong because one of the argument was used twice, and in
6149 one call a function was used. (As it happened, calling the function twice
6150 did not affect the overall behaviour.) The macro has been replaced by a
6151 function that can join an arbitrary number of extra strings onto a growing
6154 (c) The code for expansion conditions now uses a table and a binary chop
6155 instead of a serial search (which was left over from when there were very
6156 few conditions). Also, it now recognizes conditions like "pam" even when
6157 the relevant support is not compiled in: a suitably worded error message is
6158 given if an attempt is made to use such a condition.
6160 11. Added ${time_interval:xxxxx}.
6162 12. A bug was causing one of the ddress fields not to be passed back correctly
6163 from remote delivery subprocesses. The field in question was not being
6164 subsequently used, so this caused to problems in practice.
6166 13. Added new log selectors queue_time and deliver_time.
6168 14. Might have fixed a bug in maildirsizefile handling that threw up
6169 "unexpected character" debug warnings, and recalculated the data
6170 unnecessarily. In any case, I expanded the warning message to give more
6173 15. Added the message "Restricted characters in address" to the statements in
6174 the default ACL that block characters like @ and % in local parts.
6176 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
6177 Three changes have been made:
6179 (a) There was a serious bug; a negative response to MAIL caused the whole
6180 recipient domain to be cached as invalid, thereby blocking all messages
6181 to all local parts at the same domain, from all senders. This bug has
6182 been fixed. The domain is no longer cached after a negative response to
6183 MAIL if the sender used is not empty.
6185 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
6188 (c) A new callout option, "use_sender" has been added for people who want
6189 the modified behaviour.
6195 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
6198 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
6199 indeed breaks things for older releases.
6201 3. Added additional logging to the case where there is a problem reading data
6202 from a filter that is running in a subprocess using a pipe, in order to
6203 try to track down a specific problem.
6205 4. Testing facility fudge: when running in the test harness and attempting
6206 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
6207 getting "No route to host". Convert this to a timeout.
6209 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
6212 6. Some OS don't have socklen_t but use size_t instead. This affects the
6213 fifth argument of getsockopt() amongst other things. This is now
6214 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
6215 can be set for individual OS. I have set it for SunOS5, OSF1, and
6216 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
6217 some earlier ones do not.
6219 7. Change 4.30/15 was not doing the test caselessly.
6221 8. The standard form for an IPv6 address literal was being rejected by address
6222 parsing in, for example, MAIL and RCPT commands. An example of this kind of
6223 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6224 this, as well as the form without the "IPv6" on the front (but only when
6225 address literals are enabled, of course).
6227 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
6229 10. Exim crashed if a message with an empty sender address specified by -f
6230 encountered a router with an errors_to setting. This could be provoked only
6231 by a command such as
6235 where an empty string was supplied; "<>" did not hit this bug.
6237 11. Installed PCRE release 4.5.
6239 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
6240 remained set. It is now erased.
6242 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
6243 times from message ids (which are base 36 rather than the normal 62).
6245 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
6246 were being counted as actual protocol errors, and logged if the log
6247 selector +smtp_protocol_error was set. One cannot be perfect in this test,
6248 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
6249 and DATA following a set of rejected RCPTs do not count as protocol errors.
6250 In other words, Exim assumes they were pipelined, though this may not
6251 actually be the case. Of course, in all cases the client gets an
6252 appropriate error code.
6254 15. If a lookup fails in an ACL condition, a message about the failure may
6255 be available; it is used if testing the ACL cannot continue, because most
6256 such messages specify what the cause of the deferral is. However, some
6257 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
6258 that caused an old message to be retained and used if a later statement
6259 caused a defer, replacing the real cause of the deferral.
6261 16. If an IP address had so many PTR records that the DNS lookup buffer
6262 was not large enough to hold them, Exim could crash while trying to process
6263 the truncated data. It now detects and logs this case.
6265 17. Further to 4.21/58, another change has been made: if (and only if) the
6266 first line of a message (the first header line) ends with CRLF, a bare LF
6267 in a subsequent header line has a space inserted after it, so as not to
6268 terminate the header.
6270 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
6271 unnecessarily, used atoi() instead of strtol(), and didn't check the
6272 termination when getting file sizes from file names by regex.
6274 19. Completely re-implemented the support for maildirsize files, in the light
6275 of a number of problems with the previous contributed implementation
6276 (4.30/29). In particular:
6278 . If the quota is zero, the maildirsize file is maintained, but no quota is
6281 . If the maildir directory does not exist, it is created before any attempt
6282 to write a maildirsize file.
6284 . The quota value in the file is just a cache; if the quota is changed in
6285 the transport, the new value overrides.
6287 . A regular expression is available for excluding directories from the
6290 20. The autoreply transport checks the characters in options that define the
6291 message's headers; it allows continued headers, but it was checking with
6292 isspace() after an embedded newline instead of explicitly looking for a
6295 21. If all the "regular" hosts to which an address was routed had passed their
6296 expiry times, and had not reached their retry times, the address was
6297 bounced, even if fallback hosts were defined. Now Exim should go on to try
6300 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
6301 equivalent code in the SMTP transport. Some hosts send humungous responses
6302 to HELO/EHLO, more than 1024 it seems.
6304 23. Refactoring: code in filter.c used (void *) for "any old type" but this
6305 gives compiler warnings in some environments. I've now done it "properly",
6308 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
6309 (because of problems with the built-in one) was declared to return uschar *
6310 instead of char *, causing compiler failure.
6312 25. Fixed a file descriptor leak when processing alias/forward files.
6314 26. Fixed a minor format string issue in dbfn.c.
6316 27. Typo in exim.c: ("dmbnz" for "dbmnz").
6318 28. If a filter file refered to $h_xxx or $message_headers, and the headers
6319 contained RFC 2047 "words", Exim's memory could, under certain conditions,
6322 29. When a sender address is verified, it is cached, to save repeating the test
6323 when there is more than one recipient in a message. However, when the
6324 verification involves a callout, it is possible for different callout
6325 options to be set for different recipients. It is too complicated to keep
6326 track of this in the cache, so now Exim always runs a verification when a
6327 callout is required, relying on the callout cache for the optimization.
6328 The overhead is duplication of the address routing, but this should not be
6331 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
6332 to be verified with callout=postmaster, and the main callout worked but the
6333 postmaster check failed, the verification correctly failed. However, if a
6334 subsequent RCPT command asked for sender verification *without* the
6335 postmaster check, incorrect caching caused this verification also to fail,
6338 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
6339 it was not caching the DNS options (qualify_single, search_parents) that
6340 were used when the lookup failed. A subsequent lookup with different
6341 options therefore always gave the same answer, though there were cases
6342 where it should not have. (Example: a "domains = !$mx_any" option on a
6343 dnslookup router: the "domains" option is always processed without any
6344 widening, but the router might have qualify_single set.) Now Exim uses the
6345 cached value only when the same options are set.
6347 32. Added John Jetmore's "exipick" utility to the distribution.
6349 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
6350 than a timeout (e.g. a certificate is required, and is not provided), an
6351 Exim server now closes the connection immediately. Previously it waited for
6352 the client to close - but if the client is SSL, it seems that they each
6353 wait for each other, leading to a delay before one of them times out.
6355 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
6356 maintained 0.8.x compatibility because I don't think many are using it, and
6357 it is clearly obsolete.
6359 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
6362 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
6363 client certificate was expired. A simple patch fixes this, though I don't
6364 understand the full logic of why the verify callback is called multiple
6367 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
6368 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
6369 which causes problems with some clients (such as the Certicom SSL Plus
6370 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
6371 disables the coutermeasure allowing Eudora to connect."
6373 38. Exim was not checking that a write() to a log file succeeded. This could
6374 lead to Bad Things if a log got too big, in particular if it hit a file
6375 size limit. Exim now panics and dies if it cannot write to a log file, just
6376 as it does if it cannot open a log file.
6378 39. Modified OS/Makefile-Linux so that it now contains
6380 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
6382 The two -D definitions ensure that Exim is compiled with large file
6383 support, which makes it possible to handle log files that are bigger than
6386 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
6387 instance) a domain was checked against a named list that involved a lookup,
6388 causing $domain_data to be set, then another domain was checked against the
6389 same list, then the first domain was re-checked, the value of $domain_data
6390 after the final check could be wrong. In particular, if the second check
6391 failed, it could be set empty. This bug probably also applied to
6394 41. The strip_trailing_dot option was not being applied to the address given
6395 with the -f command-line option.
6397 42. The code for reading a message's header from the spool was incrementing
6398 $received_count, but never initializing it. This meant that the value was
6399 incorrect (doubled) while delivering a message in the same process in which
6400 it was received. In the most common configuration of Exim, this never
6401 happens - a fresh exec is done - but it can happen when
6402 deliver_drop_privilege is set.
6404 43. When Exim logs an SMTP synchronization error - client data sent too soon -
6405 it now includes up to 150 characters of the unexpected data in the log
6408 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
6409 and building data strings. The size of both of these buffers was 10 000
6410 bytes - far larger than anybody would *ever* want, thought I. Needless to
6411 say, somebody hit the limit. I have increased the maximum line length to
6412 20 000 and the maximum data length of concatenated lines to 100 000. I have
6413 also fixed two bugs, because there was no checking on these buffers. Tsk,
6414 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
6415 buffer is too small.
6417 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
6418 lsearch lookups. Now it does.
6420 46. When parsing a route_list item in a manualroute router, a fixed-length
6421 buffer was used for the list of hosts. I made this 1024 bytes long,
6422 thinking that nobody would ever have a list of hosts that long. Wrong.
6423 Somebody had a whole pile of complicated expansion conditions, and the
6424 string was silently truncated, leading to an expansion error. It turns out
6425 that it is easier to change to an unlimited length (owing to other changes
6426 that have happened since this code was originally written) than to build
6427 structure for giving a limitation error. The length of the item that
6428 expands into the list of hosts is now unlimited.
6430 47. The lsearch lookup could not handle data where the length of text line was
6431 more than 4095 characters. Such lines were truncated, leading to shortened
6432 data being returned. It should now handle lines of any length.
6434 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
6435 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
6438 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
6439 replacing strings such as BIN_DIRECTORY in a source file: the replacement
6440 no longer happens in comment lines. A list of replacements is now placed
6441 at the head of all of the source files, except those whose only change is
6442 to replace PERL_COMMAND in the very first #! line.
6444 50. Replaced the slow insertion sort in queue.c, for sorting the list of
6445 messages on the queue, with a bottom-up merge sort, using code contributed
6446 by Michael Haardt. This should make operations like -bp somewhat faster on
6447 large queues. It won't affect queue runners, except when queue_run_in_order
6450 51. Installed eximstats 1.31 in the distribution.
6452 52. Added support for SRV lookups to the dnslookup router.
6454 53. If an ACL referred to $message_body or $message_body_end, the value was not
6455 reset for any messages that followed in the same SMTP session.
6457 54. The store-handling optimization for building very long strings was not
6458 differentiating between the different store pools. I don't think this
6459 actually made any difference in practice, but I've tidied it.
6461 55. While running the routers to verify a sender address, $sender_address
6462 was still set to the sender address. This is wrong, because when routing to
6463 send a bounce to the sender, it would be empty. Therefore, I have changed
6464 it so that, while verifying a sender address, $sender_address is set to <>.
6465 (There is no change to what happens when verifying a recipient address.)
6467 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
6468 target A or AAAA records (if not already returned) without resetting the
6469 qualify_single or search_parents options of the DNS resolver. These are
6470 inappropriate in this case because the targets of MX and SRV records must
6471 be FQDNs. A broken DNS record could cause trouble if it happened to have a
6472 target that, when qualified, matched something in the local domain. These
6473 two options are now turned off when doing these lookups.
6475 57. It seems that at least some releases of Reiserfs (which does not have the
6476 concept of a fixed number of inodes) returns zero and not -1 for the
6477 number of available inodes. This interacted badly with check_spool_inodes,
6478 which assumed that -1 was the "no such thing" setting. What I have done is
6479 to check that the total number of inodes is greater than zero before doing
6480 the test of how many are available.
6482 58. When a "warn" ACL statement has a log_message modifier, the message is
6483 remembered, and not repeated. This is to avoid a lot of repetition when a
6484 message has many recipients that cause the same warning to be written.
6485 However, Exim was preserving the list of already written lines for an
6486 entire SMTP session, which doesn't seem right. The memory is now reset if a
6487 new message is started.
6489 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6490 debugging output unless log_rewrite was also set.
6492 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6493 of (int)(handle) when we know that handle contains (void *)(-1).
6495 61. The Exim daemon panic-logs an error return when it closes the incoming
6496 connection. However "connection reset by peer" seems to be common, and
6497 isn't really an error worthy of noting specially, so that particular error
6500 62. When Exim is trying to find all the local interfaces, it used to panic and
6501 die if the ioctl to get the interface flags failed. However, it seems that
6502 on at least one OS (Solaris 9) it is possible to have an interface that is
6503 included in the list of interfaces, but for which you get a failure error
6504 for this call. This happens when the interface is not "plumbed" into a
6505 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6506 failure of the "get flags" call assumes that the interface is down.
6508 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6509 makes life easier for people who are doing arithmetic on fields extracted
6510 from dates, where you often get leading zeros that should not be
6511 interpreted as octal.
6513 64. Added qualify_domain to the redirect router, to override the global
6516 65. If a pathologically long header line contained very many addresses (the
6517 report of this problem mentioned 10 000) and each of them was rewritten,
6518 Exim could use up a very large amount of memory. (It kept on making new
6519 copies of the header line as it rewrote, and never released the old ones.)
6520 At the expense of a bit more processing, the header rewriting function has
6521 been changed so that it no longer eats memory in this way.
6523 66. The generation of the Received: header has been moved from the time that a
6524 message starts to be received, to the time that it finishes. The timestamp
6525 in the Received: header should now be very close to that of the <= log
6526 line. There are two side-effects of this change:
6528 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6529 logged header lines no longer include the local Received: line, because
6530 it has not yet been created. The same applies to a copy of the message
6531 that is returned to a non-SMTP sender when a message is rejected.
6533 (b) When a filter file is tested using -bf, no additional Received: header
6534 is added to the test message. After some thought, I decided that this
6537 This change does not affect the value of $received_for. It is still set
6538 after address rewriting, but before local_scan() is called.
6540 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6542 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6543 gave an unhelpful panic error message, and a defer error. I have managed to
6544 change this behaviour so that it now rejects any supplied certificate,
6545 which seems right, as the list of acceptable certificates is empty.
6547 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6548 gave an unhelpful defer error. I have not managed to make this reject any
6549 supplied certificates, but the error message it gives is "no certificate
6550 supplied", which is not helpful.
6552 70. exigrep's output now also includes lines that are not associated with any
6553 message, but which match the given pattern. Implemented by a patch from
6554 Martin Sluka, which also tidied up the Perl a bit.
6556 71. Recipient callout verification, like sender verification, was using <> in
6557 the MAIL FROM command. This isn't really the right thing, since the actual
6558 sender may affect whether the remote host accepts the recipient or not. I
6559 have changed it to use the actual sender in the callout; this means that
6560 the cache record is now keyed on a recipient/sender pair, not just the
6561 recipient address. There doesn't seem to be a real danger of callout loops,
6562 since a callout by the remote host to check the sender would use <>.
6563 [SEE ABOVE: changed after hitting problems.]
6565 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6566 temporary errors. However, in the case of such a code being given after
6567 the end of a data transmission (i.e. after ".") Exim was failing to write
6568 a retry record for the message. (Yes, there was some broken host that was
6569 actually sending 8xx at this point.)
6571 73. An unknown lookup type in a host list could cause Exim to panic-die when
6572 the list was checked. (An example that provoked this was putting <; in the
6573 middle of a list instead of at the start.) If this happened during a DATA
6574 ACL check, a -D file could be left lying around. This kind of configuration
6575 error no longer causes Exim to die; instead it causes a defer error. The
6576 incident is still logged to the main and panic logs.
6578 74. Buglet left over from Exim 3 conversion. The message "too many messages
6579 in one connection" was written to the rejectlog but not the mainlog, except
6580 when address rewriting (yes!) was being logged.
6582 75. Added write_rejectlog option.
6584 76. When a system filter was run not as root (that is, when system_filter_user
6585 was set), the values of the $n variables were not being returned to the
6586 main process; thus, they were not subsequently available in the $sn
6589 77. Added +return_path_on_delivery log selector.
6591 78. A connection timeout was being treated differently from recipients deferred
6592 when testing hosts_max_try with a message that was older than the host's
6593 retry timeout. (The host should not be counted, thus allowing all hosts to
6594 be tried at least once before bouncing.) This may have been the cause of an
6595 occasionally reported bug whereby a message would remain on the queue
6596 longer than the retry timeout, but would be bounced if a delivery was
6597 forced. I say "may" because I never totally pinned down the problem;
6598 setting up timeout/retry tests is difficult. See also the next item.
6600 79. The ultimate address timeout was not being applied to errors that involved
6601 a combination of host plus message (for example, a timeout on a MAIL
6602 command). When an address resolved to a number of possible hosts, and they
6603 were not all tried for each delivery (e.g. because of hosts_max_try), a
6604 message could remain on the queue longer than the retry timeout.
6606 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6609 81. Fixed an obscure SMTP outgoing bug which required at least the following
6610 conditions: (a) there was another message waiting for the same server;
6611 (b) the server returned 5xx to all RCPT commands in the first message so
6612 that the message was not completed; (c) the server dropped the connection
6613 or gave a negative response to the RSET that Exim sends to abort the
6614 transaction. The observed case was a dropped connection after DATA that had
6615 been sent in pipelining mode. That is, the server had advertised PIPELINING
6616 but was not implementing it correctly. The effect of the bug was incorrect
6617 behaviour, such as trying another host, and this could lead to a crash.
6623 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6624 and daemon.c were passed as pointers to ints; they should have been
6625 pointers to socklen_t variables (which are typically unsigned ints).
6627 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6630 3. Fixed a really odd bug that affected only the testing scheme; patching a
6631 certain fixed string in the binary changed the value of another string that
6632 happened to be identical to the end of the original first string.
6634 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6635 name", it returns that address as the IP address. On some operating
6636 systems (e.g. Solaris), it also passes back the IP address string as the
6637 "host name". However, on others (e.g. Linux), it passes back an empty
6638 string. Exim wasn't checking for this, and was changing the host name to an
6639 empty string, assuming it had been canonicalized.
6641 5. Although rare, it is permitted to have more than one PTR record for a given
6642 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6643 all the names associated with an address, because they do in Solaris.
6644 However, it seems that they do not in Linux for data that comes from the
6645 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6646 I found this out when I moved to a new Linux workstation and tried to run
6647 the Exim test suite.
6649 To get round this problem I have changed the code so that it now does its
6650 own call to the DNS to look up PTR records when searching for a host name.
6651 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6652 addresses that are only in /etc/hosts are still found.
6654 This behaviour is, however, controlled by an option called host_lookup_
6655 order, which defaults to "bydns:byaddr". If people want to use the other
6656 order, or indeed, just use one or the other means of lookup, they can
6657 specify it in this variable.
6659 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6660 some operating systems, this comes back from gethostbyaddr() as an empty
6661 string, and this is what Exim used to test for. However, it seems that in
6662 other systems, "." is yielded. Exim now tests for this case too.
6664 7. The values of check_spool_space and check_log_space are now held internally
6665 as a number of kilobytes instead of an absolute number of bytes. If a
6666 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6667 kilobyte. This means that much larger values can be stored.
6669 8. Exim monitor: an attempt to get the action menu when not actually pointing
6670 at a message produces an empty menu entitled "No message selected". This
6671 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6672 no entries in it ("Shell widget menu has zero width and/or height"). So I
6673 have added a single, blank menu entry in this case.
6675 9. Added ${quote_local_part.
6677 10. MIME decoding is now applied to the contents of Subject: header lines when
6680 11. Now that a reference to $sender_host_address automatically causes a reverse
6681 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6682 host lookup before query-style lookups in lists that might use this
6683 variable. This has therefore been abolished, and the "net-" prefix is no
6684 longer necessary for query-style lookups.
6686 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6687 have been a typo for LFLAGS, so it has been changed.
6689 13. The install script calls Exim with "-C /dev/null" in order to find the
6690 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6691 to be output. However, since Exim outputs its version number before the
6692 error, it didn't break the script. It just looked ugly. I fixed this by
6693 always allowing "-C /dev/null" if the caller is root.
6695 14. Ignore overlarge ACL variable number when reading spool file - insurance
6696 against a later release with more variables having written the file.
6698 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6699 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6700 this, as well as the form without the "IPv6" on the front.
6702 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6703 OS/Makefile-Darwin file.
6705 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6706 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6708 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6709 variable. However, some OS use ints for inodes. I've added cast to long int
6710 to get rid of the compiler warning.
6712 19. I had forgotten to lock out "/../" in configuration file names when
6713 ALT_CONFIG_PREFIX was set.
6715 20. Routers used for verification do not need to specify transports. However,
6716 if such a router generated a host list, and callout was configured, Exim
6717 crashed, because it could not find a port number from the (non-existent)
6718 transport. It now assumes port 25 in this circumstance.
6720 21. Added the -t option to exigrep.
6722 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6723 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6724 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6725 them is set, LOOKUP_LSEARCH is forced.
6727 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6729 24. Added sender and host information to the "rejected by local_scan()" log
6730 line; previously there was no indication of these.
6732 25. Added .include_if_exists.
6734 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6735 Linux. It turns out that not all file systems support this. Apparently some
6736 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6737 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6738 not supported on the file descriptor, is now ignored when Exim is trying to
6739 sync a directory. This applies only to Linux.
6741 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6743 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6744 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6747 29. Added support for maildirsize files from supplied patch (modified a bit).
6749 30. The use of :fail: followed by an empty string could lead Exim to respond to
6750 sender verification failures with (e.g.):
6752 550 Verification failed for <xxx>
6753 550 Sender verify failed
6755 where the first response line was missing the '-' that indicates it is not
6756 the final line of the response.
6758 31. The loop for finding the name of the user that called Exim had a hardwired
6759 limit of 10; it now uses the value of finduser_retries, which is used for
6760 all other user lookups.
6762 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6765 33. Exim was neglecting to zero errno before one call of strtol() when
6766 expanding a string and expecting an integer value. On some systems this
6767 resulted in spurious "integer overflow" errors. Also, it was casting the
6768 result into an int without checking.
6770 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6771 did not work. The code looks as if it has *never* worked, though it appears
6772 to have been documented since at least release 1.62. I have made it work.
6774 35. The "timeout_DNS" error in retry rules, also documented since at least
6775 1.62, also never worked. As it isn't clear exactly what this means, and
6776 clearly it isn't a major issue, I have abolished the feature by treating it
6777 as "timeout", and writing a warning to the main and panic logs.
6779 36. The display of retry rules for -brt wasn't always showing the error code
6782 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6783 timeout_connect_A, timeout_connect_MX.
6785 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6786 to the empty sender.
6788 39. The daemon was not analysing the content of -oX till after it had closed
6789 stderr and disconnected from the controlling terminal. This meant that any
6790 syntax errors were only noted on the panic log, and the return code from
6791 the command was 0. By re-arranging the code a little, I've made the
6792 decoding happen first, so such errors now appear on stderr, and the return
6793 code is 1. However, the actual setting up of the sockets still happens in
6794 the disconnected process, so errors there are still only recorded on the
6797 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6798 connections (as happens on some IP stacks) was logged at start up time as
6799 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6800 it from "IPv6 and IPv4", which means that two separate sockets are being
6803 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6804 says whether AF_INET or AF_INET6 was passed as an argument.
6806 42. Exiwhat output was messed up when time zones were included in log
6809 43. Exiwhat now gives more information about the daemon's listening ports,
6810 and whether -tls-on-connect was used.
6812 44. The "port" option of the smtp transport is now expanded.
6814 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6815 silently ignored. Now an error message is written to the main and panic
6818 46. There's a new ACL modifier called "logwrite" which writes to a log file
6819 as soon as it is encountered.
6821 47. Added $local_user_uid and $local_user_gid at routing time.
6823 48. Exim crashed when trying to verify a sender address that was being
6826 49. Exim was recognizing only a space character after ".include". It now also
6827 recognizes a tab character.
6829 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6830 extracting the relevant information from the specification. The man page no
6831 longer contains scrambled data for the -d option, and I've added a section
6832 at the front about calling Exim under different names.
6834 51. Added "extra_headers" argument to the "mail" command in filter files.
6836 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6839 53. Installed eximstats 1.29.
6841 54. Added transport_filter_timeout as a generic transport option.
6843 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6844 Cc: header lines. This was required by RFC 822, but it not required by RFC
6847 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6848 incoming messages that did not have them. Now it does so only if the
6849 message originates locally, that is, if there is no associated remote host
6850 address. When Resent- header lines are present, this applies to the Resent-
6851 lines rather than the non-Resent- lines.
6853 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6854 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6856 58. Messages for configuration errors now include the name of the main
6857 configuration file - useful now that there may be more than one file in a
6858 list (.included file names were always shown).
6860 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6861 for those rare installations that do not start the daemon as root or run it
6862 setuid root. I've cut out the call to initgroups() if the daemon is not
6865 60. The Exim user and group can now be bound into the binary as text strings
6866 that are looked up at the start of Exim's processing.
6868 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6870 62. Added $mailstore_basename variable.
6872 63. Installed patch to sieve.c from Michael Haardt.
6874 64. When Exim failed to open the panic log after failing to open the main log,
6875 the original message it was trying to log was written to stderr and debug
6876 output, but if they were not available (the usual case in production), it
6877 was lost. Now it is written to syslog before the two lines that record the
6878 failures to open the logs.
6880 65. Users' Exim filters run in subprocesses under the user's uid. It is
6881 possible for a "deliver" command or an alias in a "personal" command to
6882 provoke an address rewrite. If logging of address rewriting is configured,
6883 this fails because the process is not running as root or exim. There may be
6884 a better way of dealing with this, but for the moment (because 4.30 needs
6885 to be released), I have disabled address rewrite logging when running a
6886 filter in a non-root, non-exim process.
6892 1. The buildconfig auxiliary program wasn't quoting the value set for
6893 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6894 not defined. This bug was masked in 4.22 by the effect that was fixed in
6897 2. Some messages that were rejected after a message id was allocated were
6898 shown as "incomplete" by exigrep. It no longer does this for messages that
6899 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6901 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6902 have allow_domain_literals set, the ID did not get logged in the <= line.
6903 Domain literals are now always recognized in Message-ID: header lines.
6905 4. The first argument for a ${extract expansion item is the key name or field
6906 number. Leading and trailing spaces in this item were not being ignored,
6907 causing some misleading effects.
6909 5. When deliver_drop_privilege was set, single queue runner processes started
6910 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6911 same command in the process it spins off) were not dropping privilege.
6913 6. When the daemon running as "exim" started a queue runner, it always
6914 re-executed Exim in the spun-off process. This is a waste of effort when
6915 deliver_drop_privilege is set. The new process now just calls the
6916 queue-runner function directly.
6922 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6925 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6926 ignored. Though the use of -oP was forcing the writing of a pid file, it
6927 was always written to the default place.
6929 3. If the message "no IP address found for host xxxx" is generated during
6930 incoming verification, it is now followed by identification of the incoming
6931 connection (so you can more easily find what provoked it).
6933 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6935 5. Added some features to "harden" Exim a bit more against certain attacks:
6937 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6938 be put in Local/Makefile. This is like the never_users runtime option,
6939 but it cannot be overridden. The default setting is "root".
6941 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6942 prefix string with which any file named in a -C command line option
6945 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6946 is retained for -C and -D only if the caller of Exim is root. Without
6947 it, the exim user may also use -C and -D and retain privilege.
6949 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6950 command line option is disabled.
6952 6. Macro names set by the -D option must start with an upper case letter, just
6953 like macro names defined in the configuration file.
6955 7. Added "dereference=" facility to LDAP.
6957 8. Two instances of the typo "uknown" in the source files are fixed.
6959 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6960 the Configure-Makefile script screwed up while processing it.
6962 10. Incorporated PCRE 4.4.
6964 11. The SMTP synchronization check was not operating right at the start of an
6965 SMTP session. For example, it could not catch a HELO sent before the client
6966 waited for the greeting. There is now a check for outstanding input at the
6967 point when the greeting is written. Because of the duplex, asynchronous
6968 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6969 way, but not yet received, when the check is performed.
6971 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6972 on TCP/IP sockets, because this apparently causes some broken clients to
6975 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6976 unchanged) from the Cygwin maintainer.
6978 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6979 was supported instead of testing for mbx. Effectively a typo.
6981 15. The spa authenticator server code was not checking that the input it
6982 received was valid base64.
6984 16. The debug output line for the "set" modifier in ACLs was not showing the
6985 name of the variable that was being set.
6987 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6989 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6990 lookup on its own. Something else was needed to trigger the lookup. For
6991 example, a match in host_lookup or the need for a host name in a host list.
6992 Now, if $sender_host_name is referenced and the host name has not yet been
6993 looked up, a lookup is performed. If the lookup fails, the variable remains
6994 empty, and $host_lookup_failed is set to "1".
6996 19. Added "eqi" as a case-independent comparison operator.
6998 20. The saslauthd authentication condition could segfault if neither service
6999 nor realm was specified.
7001 21. If an overflowing value such as "2048M" was set for message_size_limit, the
7002 error message that was logged was misleading, and incoming SMTP
7003 connections were dropped. The message is now more accurate, and temporary
7004 errors are given to SMTP connections.
7006 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
7007 (except RSET) with a 421 error, until QUIT is received. However, it was
7008 failing to send a response to QUIT.
7010 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
7011 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
7012 ACL is now run after the helo_try_verify_hosts code.
7014 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
7015 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
7016 case-independent, so other case variants are also recognized). Apparently
7017 some systems use these upper case variants.
7019 25. If more than two messages were waiting for the same host, and a transport
7020 filter was specified for the transport, Exim sent two messages over the
7021 same TCP/IP connection, and then failed with "socket operation on non-
7022 socket" when it tried to send the third.
7024 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
7026 27. The extern definition of crypt16() in expand.c was not being excluded when
7027 the OS had its own crypt16() function.
7029 28. Added bounce_return_body as a new option, and bounce_return_size_limit
7030 as a preferred synonym for return_size_limit, both as an option and as an
7033 29. Added LIBS=-liconv to OS/Makefile-OSF1.
7035 30. Changed the default configuration ACL to relax the local part checking rule
7036 for addresses that are not in any local domains. For these addresses,
7037 slashes and pipe symbols are allowed within local parts, but the sequence
7038 /../ is explicitly forbidden.
7040 31. SPA server authentication was not clearing the challenge buffer before
7043 32. log_message in a "warn" ACL statement was writing to the reject log as
7044 well as to the main log, which contradicts the documentation and doesn't
7045 seem right (because no rejection is happening). So I have stopped it.
7047 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
7048 However, I am unable to do any testing of this.
7050 34. Fixed an infelicity in the appendfile transport. When checking directories
7051 for a mailbox, to see if any needed to be created, it was accidentally
7052 using path names with one or more superfluous leading slashes; tracing
7053 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
7055 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
7056 contents are added to the log line that is written for every discarded
7057 recipient. (Previously a log_message setting was ignored.)
7059 36. The ${quote: operator now quotes the string if it is empty.
7061 37. The install script runs exim in order to find its version number. If for
7062 some reason other than non-existence or emptiness, which it checks, it
7063 could not run './exim', it was installing it with an empty version number,
7064 i.e. as "exim-". This error state is now caught, and the installation is
7067 38. An argument was missing from the function that creates an error message
7068 when Exim fails to connect to the socket for saslauthd authentication.
7069 This could cause Exim to crash, or give a corrupted message.
7071 39. Added isip, isip4, and isip6 to ${if conditions.
7073 40. The ACL variables $acl_xx are now saved with the message, and can be
7074 accessed later in routers, transports, and filters.
7076 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
7077 strings in the file are not string-expanded.
7079 42. If a MAIL command specified a SIZE value that was too large to fit into an
7080 int variable, the check against message_size_limit failed. Such values are
7081 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
7082 day this will have to be increased, but I don't think I want to be around
7083 when emails are that large.
7090 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
7091 iconv() is not standard in FreeBSD.
7093 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
7094 IPv6 enabled. The observed symptom was a segmentation fault on return from
7095 the function os_common_find_running_interfaces() in src/os.c.
7097 3. In the check_special_case() function in daemon.c I had used "errno" as an
7098 argument name, which causes warnings on some systems. This was basically a
7099 typo, since it was named "eno" in the comments!
7101 4. The code that waits for the clock to tick (at a resolution of some fraction
7102 of a second) so as to ensure message-id uniqueness was always waiting for
7103 at least one whole tick, when it could have waited for less. [This is
7104 almost certainly not relevant at current processor speeds, where it is
7105 unlikely to ever wait at all. But we try to future-proof.]
7107 5. The function that sleeps for a time interval that includes fractions of a
7108 second contained a race. It did not block SIGALRM between setting the
7109 timer, and suspending (a couple of lines later). If the interval was short
7110 and the sigsuspend() was delayed until after it had expired, the suspension
7111 never ended. On busy systems this could lead to processes getting stuck for
7114 6. Some uncommon configurations may cause a lookup to happen in a queue runner
7115 process, before it forks any delivery processes. The open lookup caching
7116 mechanism meant that the open file or database connection was passed into
7117 the delivery process. The problem was that delivery processes always tidy
7118 up cached lookup data. This could cause a problem for the next delivery
7119 process started by the queue runner, because the external queue runner
7120 process does not know about the closure. So the next delivery process
7121 still has data in the lookup cache. In the case of a file lookup, there was
7122 no problem because closing a file descriptor in a subprocess doesn't affect
7123 the parent. However, if the lookup was caching a connection to a database,
7124 the connection was closed, and the second delivery process was likely to
7125 see errors such as "PGSQL: query failed: server closed the connection
7126 unexpectedly". The problem has been fixed by closing all cached lookups
7127 in a queue runner before running a delivery process.
7129 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
7130 seem to have the "const" qualifier which it has on other OS. I've
7133 8. Change 4.21/2 was too strict. It is only if there are two authenticators
7134 *of the same type* (client or server) with the same public name that an
7135 error should be diagnosed.
7137 9. When Exim looked up a host name for an IP address, but failed to find the
7138 original IP address when looking up the host name (a safety check), it
7139 output the message "<ip address> does not match any IP for NULL", which was
7140 confusing, to say the least. The bug was that the host name should have
7141 appeared instead of "NULL".
7143 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
7144 user that is built into the binary, and the -C or -D options is used, root
7145 privilege is dropped before the configuration file is read. In addition,
7146 logging is switched to stderr instead of the normal log files. If the
7147 configuration then re-defines the Exim user, the unprivileged environment
7148 is probably not what is expected, so Exim logs a panic warning message (but
7151 However, if deliver_drop_privilege is set, the unprivileged state may well
7152 be exactly what is intended, so the warning has been cut out in that case,
7153 and Exim is allowed to try to write to its normal log files.
7159 1. smtp_return_error_details was not giving details for temporary sender
7160 or receiver verification errors.
7162 2. Diagnose a configuration error if two authenticators have the same public
7165 3. Exim used not to create the message log file for a message until the first
7166 delivery attempt. This could be confusing when incoming messages were held
7167 for policy or load reasons. The message log file is now created at the time
7168 the message is received, and an initial "Received" line is written to it.
7170 4. The automatically generated man page for command line options had a minor
7171 bug that caused no ill effects; however, a more serious problem was that
7172 the procedure for building the man page automatically didn't always
7173 operate. Consequently, release 4.20 contains an out-of-date version. This
7174 shouldn't happen again.
7176 5. When building Exim with embedded Perl support, the script that builds the
7177 Makefile was calling 'perl' to find its compile-time parameters, ignoring
7178 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
7180 6. The freeze_tell option was not being used for messages that were frozen on
7181 arrival, either by an ACL or by local_scan().
7183 7. Added the smtp_incomplete_transaction log selector.
7185 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
7186 was accepting AUTH without a new EHLO.
7188 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
7189 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
7192 10. Logging of TCP/IP connections (when configured) now happens in the main
7193 daemon process instead of the child process, so that the TCP/IP connection
7194 count is more accurate (but it can never be perfect).
7196 11. The use of "drop" in a nested ACL was not being handled correctly in the
7197 outer ACL. Now, if condition failure induced by the nested "drop" causes
7198 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
7199 or "require"), the connection is dropped.
7201 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
7202 that yield "discard" can now be used with an "accept" or a "discard" verb,
7203 but an error is generated for any others (because I can't see a useful way
7204 to define what should happen).
7206 13. When an ACL is read dynamically from a file (or anywhere else), the lines
7207 are now processed in the same way as lines in the Exim configuration file.
7208 In particular, continuation lines are supported.
7210 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
7212 15. Added -ti meaning -t -i.
7214 16. Check for letters, digits, hyphens, and dots in the names of dnslist
7215 domains, and warn by logging if others are found.
7217 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
7218 returned from GIFCONF when Exim is trying to find the list of interfaces on
7219 a host. The code in os.c has been modified to copy each ifreq to an aligned
7220 structure in all cases.
7222 Also, in some cases, the returned ifreq's were being copied to a 'struct
7223 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
7224 means the last couple of bytes of an IPv6 address could be chopped if the
7225 ifreq contained only a normal sockaddr (14 bytes storage).
7227 18. Named domain lists were not supported in the hosts_treat_as_local option.
7228 An entry such as +xxxx was not recognized, and was treated as a literal
7231 19. Ensure that header lines added by a DATA ACL are included in the reject log
7232 if the ACL subsequently rejects the message.
7234 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
7235 MD5 (which is deprecated).
7237 21. When testing a filter file using -bf, Exim was writing a message when it
7238 took the sender from a "From " line in the message, but it was not doing so
7239 when it took $return_path from a Return-Path: header line. It now does.
7241 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
7242 with a valid header line field name (a series of printing characters
7243 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
7245 23. Changed "disc" in the source to "disk" to conform to the documentation and
7246 the book and for uniformity.
7248 24. Ignore Sendmail's -Ooption=value command line item.
7250 25. When execve() failed while trying to run a command in a pipe transport,
7251 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
7252 could be confused with a return value of 69 from the command itself. This
7253 has been changed to 127, the value the shell returns if it is asked to run
7254 a non-existent command. The wording for the related log line suggests a
7255 non-existent command as the problem.
7257 26. If received_header_text expands to an empty string, do not add a Received:
7258 header line to the message. (Well, it adds a token one on the spool, but
7259 marks it "old" so that it doesn't get used or transmitted.)
7261 27. Installed eximstats 1.28 (addition of -nt option).
7263 28. There was no check for failure on the call to getsockname() in the daemon
7264 code. This can fail if there is a shortage of resources on the system, with
7265 ENOMEM, for example. A temporary error is now given on failure.
7267 29. Contrary to the C standard, it seems that in some environments, the
7268 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
7269 program. Exim now does this explicitly; it affects the formatting of
7270 timestamps using strftime().
7272 30. If exiqsumm was given junk data, it threw up some uninitialized variable
7273 complaints. I've now initialized all the variables, to avoid this.
7275 32. Header lines added by a system filter were not being "seen" during
7276 transport-time rewrites.
7278 33. The info_callback() function passed to OpenSSL is set up with type void
7279 (*)(SSL *, int, int), as described somewhere. However, when calling the
7280 function (actually a macro) that sets it up, the type void(*)() is
7281 expected. I've put in a cast to prevent warnings from picky compilers.
7283 34. If a DNS black list lookup found a CNAME record, but there were no A
7284 records associated with the domain it pointed at, Exim crashed.
7286 35. If a DNS black list lookup returned more than one A record, Exim ignored
7287 all but the first. It now scans all returned addresses if a particular IP
7288 value is being sought. In this situation, the contents of the
7289 $dnslist_value variable are a list of all the addresses, separated by a
7292 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
7293 to accept a host name and all its aliases if the forward lookup for any of
7294 them yielded the IP address of the incoming connection. Now it accepts only
7295 those names whose forward lookup yields the correct IP address. Any other
7296 names are discarded. This closes a loophole whereby a rogue DNS
7297 administrator could create reverse DNS records to break through a
7298 wildcarded host restriction in an ACL.
7300 37. If a user filter or a system filter that ran in a subprocess used any of
7301 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
7302 the wrong values were passed to the pipe command ($thisaddress had the
7303 value of $0, $0 had the value of $1, etc). This bug was introduced by
7304 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
7306 38. Improved the line breaking for long SMTP error messages from ACLs.
7307 Previously, if there was no break point between 40 and 75 characters, Exim
7308 left the rest of the message alone. Two changes have been made: (a) I've
7309 reduced the minimum length to 35 characters; (b) if it can't find a break
7310 point between 35 and 75 characters, it looks ahead and uses the first one
7311 that it finds. This may give the occasional overlong line, but at least the
7312 remaining text gets split now.
7314 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
7315 file descriptors might be low, and that setting 1000 would always raise it.
7316 It turns out that in some environments, the limit is already over 1000 and
7317 that lowering it causes trouble. So now Exim takes care not to decrease it.
7319 40. When delivering a message, the value of $return_path is set to $sender_
7320 address at the start of routing (routers may change the value). By an
7321 oversight, this default was not being set up when an address was tested by
7322 -bt or -bv, which affected the outcome if any router or filter referred to
7325 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
7326 log only when "message" or "log_message" are set. However, if one of the
7327 conditions was an address verification, or a call to a nested ACL, the
7328 messages generated by the underlying test were being passed through. This
7329 no longer happens. The underlying message is available in $acl_verify_
7330 message for both "message" and "log_message" expansions, so it can be
7331 passed through if needed.
7333 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
7334 new expansion $bh_ to give the encoded byte string without charset
7335 translation. Translation happens only if iconv() is available; HAVE_ICONV
7336 indicates this at build time. HEADERS_CHARSET gives the charset to
7337 translate to; headers_charset can change it in the configuration, and
7338 "headers charset" can change it in an individual filter file.
7340 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
7341 that creates RFC 2047 encoded "words" labels them as that charset instead
7342 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
7343 expansion operator; (ii) when Exim creates a From: line for a local
7344 message; (iii) when a header line is rewritten to include a "phrase" part.
7346 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
7347 buggy, causing it to skip the first lines of messages whose message ID
7348 ended in 'D'. This would not have bitten before Exim release 4.14, because
7349 message IDs were unlikely to end in 'D' before then. The effect was to have
7350 incorrect size information for certain domains.
7352 45. #include "config.h" was missing at the start of the crypt16.c module. This
7353 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
7356 46. If there was a timeout during a "random" callout check, Exim treated it as
7357 a failure of the random address, and carried on sending RSET and the real
7358 address. If the delay was just some slowness somewhere, the response to the
7359 original RCPT would be taken as a response to RSET and so on, causing
7360 mayhem of various kinds.
7362 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
7363 when I implemented it. It didn't allow for the fact that some option values
7364 may legitimately be negative (e.g. size_addition), and it didn't even do
7365 the right test for positive values.
7367 48. Domain names in DNS records are case-independent. Exim always looks them up
7368 in lower case. Some resolvers return domain names in exactly the case they
7369 appear in the zone file, that is, they may contain uppercase letters. Not
7370 all resolvers do this - some return always lower case. Exim was treating a
7371 change of case by a resolver as a change of domain, similar to a widening
7372 of a domain abbreviation. This triggered its re-routing code and so it was
7373 trying to route what was effectively the same domain again. This normally
7374 caused routing to fail (because the router wouldn't handle the domain
7375 twice). Now Exim checks for this case specially, and just changes the
7376 casing of the domain that it ultimately uses when it transmits the message
7379 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
7382 50. If a filter generated a file delivery with a non-absolute name (possible if
7383 no home directory exists for the router), the forbid_file option was not
7386 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
7387 the existing equality matching.
7389 52. Exim was using ints instead of ino_t variables in some places where it was
7390 dealing with inode numbers.
7392 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
7393 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
7394 called TMPDIR, and if it finds it is different, it changes its value.
7396 54. The smtp_printf() function is now made available to local_scan() so
7397 additional output lines can be written before returning. There is also an
7398 smtp_fflush() function to enable the detection of a dropped connection.
7399 The variables smtp_input and smtp_batched_input are exported to
7402 55. Changed the default runtime configuration: the message "Unknown user"
7403 has been removed from the ACL, and instead placed on the localuser router,
7404 using the cannot_route_message feature. This means that any verification
7405 failures that generate their own messages won't get overridden. Similarly,
7406 the "Unrouteable address" message that was in the ACL for unverifiable
7407 relay addresses has also been removed.
7409 56. Added hosts_avoid_esmtp to the smtp transport.
7411 57. The exicyclog script was not checking for the esoteric option
7412 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
7413 will work only if exicyclog is run under the appropriate euid.
7415 58. Following a discussion on the list, the rules by which Exim recognises line
7416 endings on incoming messages have been changed. The -dropcr and drop_cr
7417 options are now no-ops, retained only for backwards compatibility. The
7418 following line terminators are recognized: LF CRLF CR. However, special
7419 processing applies to CR:
7421 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
7422 nor a local message in the state where . is a terminator.
7424 (ii) If a bare CR is encountered in a header line, an extra space is added
7425 after the line terminator so as not to end the header. The reasoning
7426 behind this is that bare CRs in header lines are most likely either
7427 to be mistakes, or people trying to play silly games.
7429 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
7430 was being incorrectly given as 18 bytes larger than it should have been.
7431 This is a VOB (very old bug).
7433 60. This may never have affected anything current, but just in case it has:
7434 When the local host is found other than at the start of a list of hosts,
7435 the local host, those with the same MX, and any that follow, are discarded.
7436 When the list in question was part of a longer list of hosts, the following
7437 hosts (not currently being processed) were also being discarded. This no
7438 longer happens. I'm not sure if this situation could ever has previously
7441 61. Added the "/MX" feature to lists of hosts in the manualroute and query
7444 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
7445 header. This is something that is recommended in a new Internet Draft, and
7446 is something that is documented as being done by Sendmail. There are two
7447 possible values. For messages generated by the autoreply transport, Exim
7450 Auto-Submitted: auto-replied
7452 whereas for all other generated messages (e.g. bounces) it adds
7454 Auto-Submitted: auto-generated
7456 63. The "personal" condition in filters now includes a test for the
7457 Auto-Submitted: header. If it contains the string "auto-" the message it
7458 not considered personal.
7460 64. Added rcpt_include_affixes as a generic transport option.
7462 65. Added queue_only_override (default true).
7464 66. Added the syslog_duplication option.
7466 67. If what should have been the first header line of a message consisted of
7467 a space followed by a colon, Exim was mis-interpreting it as a header line.
7468 It isn't of course - it is syntactically invalid and should therefore be
7469 treated as the start of the message body. The misbehaviour could have
7470 caused a number of strange effects, including loss of data in subsequent
7471 header lines, and spool format errors.
7473 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
7474 client host had authenticated. This control can now be exercised by an ACL
7475 for more flexibility.
7477 69. By default, callouts do not happen when testing with -bh. There is now a
7478 variant, -bhc, which does actually run the callout code, including
7479 consulting and updating the callout cache.
7481 70. Added support for saslauthd authentication, courtesy of Alexander
7484 71. If statvfs() failed on the spool or log directories while checking their
7485 size for availability, Exim confusingly gave the error "space shortage".
7486 Furthermore, in debugging mode it crashed with a floating point exception.
7487 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7488 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7489 serious problem, Exim now writes to the main and panic logs when this
7490 happens, with details of the failure. It then refuses to accept the
7491 incoming message, giving the message "spool directory problem" or "log
7492 directory problem" with a 421 code for SMTP messages.
7494 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7495 0, 1, and 2 exist, because some OS complain for execs without them (see
7496 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7497 descriptors. However, the code omitted to check that the open succeeded,
7498 causing mysterious errors if for some reason the permissions on /dev/null
7499 got screwed. Now Exim writes a message to the main and panic logs, and
7500 bombs out if it can't open /dev/null.
7502 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7503 interact so that it is all more flexible. It is supposed to remain
7504 backwards compatible. Also added extra_local_interfaces.
7506 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7507 to bomb out with an assertion failure - to the client this appears as a
7508 connection drop. This problem occurs in the part of the code that was taken
7509 from the Samba project. Fortunately, the assertion is in a very simple
7510 function, so I have fixed this by reproducing the function inline in the
7511 one place where it is called, and arranging for authentication to fail
7512 instead of killing the process with assert().
7514 75. The SPA client code was not working when the server requested OEM rather
7515 than Unicode encoding.
7517 76. Added code to make require_files with a specific uid setting more usable in
7518 the case where statting the file as root fails - usually a non-root-mounted
7519 NFS file system. When this happens and the failure is EACCES, Exim now
7520 forks a subprocess and does the per-uid checking as the relevant uid.
7522 77. Added process_log_path.
7524 78. If log_file_path was not explicitly set, a setting of check_log_space or
7525 check_log_inodes was ignored.
7527 79. If a space check for the spool or log partitions fails, the incident is now
7528 logged. Of course, in the latter case the data may get lost...
7530 80. Added the %p formatting code to string_format() so that it can be used to
7531 print addresses in debug_print(). Adjusted all the address printing in the
7532 debugging in store.c to use %p rather than %d.
7534 81. There was a concern that a line of code in smtp_in.c could overflow a
7535 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7536 initially expressed, the concern was not well-founded, because trailing
7537 spaces are removed early. However, if the trailing spaces were followed by
7538 a NULL, they did not get removed, so the overflow was possible. Two fixes
7541 (a) I re-wrote the offending code in a cleaner fashion.
7542 (b) If an incoming SMTP command contains a NULL character, it is rejected
7545 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7546 runs initgroups(), so that if the Exim user is in any additional groups,
7547 they will be used during message reception.
7553 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob