1 Change log file for Exim from version 4.21
2 ------------------------------------------
3 This document describes *changes* to previous versions, that might
4 affect Exim's operation, with an unchanged configuration file. For new
5 options, and new features, see the NewStuff file next to this ChangeLog.
11 JH/01 OpenSSL: With debug enabled output keying information sufficient, server
12 side, to decode a TLS 1.3 packet capture.
14 JH/02 OpenSSL: Suppress the sending of (stateful) TLS1.3 session tickets.
15 Previously the default library behaviour applied, sending two, each in
18 JH/03 Debug output for ACL now gives the config file name and line number for
21 JH/04 The default received_header_text now uses the RFC 8314 tls cipher clause.
23 JH/05 DKIM: ensure that dkim_domain elements are lowercased before use.
25 JH/06 Fix buggy handling of autoreply bounce_return_size_limit, and a possible
26 buffer overrun for (non-chunking) other transports.
28 JH/07 GnuTLS: Our use of late (post-handshake) certificate verification, under
29 TLS1.3, means that a server rejecting a client certificate is not visible
30 to the client until the first read of encrypted data (typically the
31 response to EHLO). Add detection for that case and treat it as a failed
32 TLS connection attempt, so that the normal retry-in-clear can work (if
35 JB/01 Bug 2375: fix expansions of 822 addresses having comments in local-part
36 and/or domain. Found and fixed by Jason Betts.
38 JH/08 Add hardening against SRV & TLSA lookups the hit CNAMEs (a nonvalid
39 configuration). If a CNAME target was not a wellformed name pattern, a
42 JH/09 Logging: Fix initial listening-on line for multiple ports for an IP when
43 the OS reports them interleaved with other addresses.
45 JH/10 OpenSSL: Fix aggregation of messages. Previously, when PIPELINING was
46 used both for input and for a verify callout, both encrypted, SMTP
47 responses being sent by the server could be lost. This resulted in
48 dropped connections and sometimes bounces generated by a peer sending
51 JH/11 Harden plaintext authenticator against a badly misconfigured client-send
52 string. Previously it was possible to cause undefined behaviour in a
53 library routine (usually a crash). Found by "zerons".
55 JH/12 Bug 2384: fix "-bP smtp_receive_timeout". Previously it returned no
58 JH/13 Bug 2386: Fix builds with Dane under LibreSSL 2.9.0 onward. Some old
59 API was removed, so update to use the newer ones.
61 JH/14 Bug 1891: Close the log file if receiving a non-smtp message, without
62 any timeout set, is taking a long time. Previously we would hang on to a
63 rotated logfile "forever" if the input was arriving with long gaps
64 (a previous attempt to fix addressed lack, for a long time, of initial
67 HS/01 Bug 2390: Use message_id for tempfile creation to avoid races in a
68 shared (NFS) environment. The length of the tempfile name is now
69 4 + 16 ("hdr.$message_exim_id") which might break on file
70 systems which restrict the file name length to lower values.
73 HS/02 Bug 2390: Use message_id for tempfile creation to avoid races in a
74 shared (NFS) environment.
76 HS/03 Bug 2392: exigrep does case sensitive *option* processing (as it
77 did for all versions <4.90). Notably -M, -m, --invert, -I may be
80 JH/15 Use unsigned when creating bitmasks in macros, to avoid build errors
81 on some platforms for bit 31.
83 JH/16 GnuTLS: rework ciphersuite strings under recent library versions. Thanks
84 to changes apparently associated with TLS1.3 handling some of the APIs
85 previously used were either nonfunctional or inappropriate. Strings
86 like TLS1.3:ECDHE_SECP256R1__RSA_PSS_RSAE_SHA256__AES_256_GCM__AEAD:256
87 and TLS1.2:ECDHE_SECP256R1__RSA_SHA256__AES_128_CBC__SHA256:128 replace
88 the previous TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256 .
89 This affects log line X= elements, the $tls_{in,out}_cipher variables,
90 and the use of specific cipher names in the encrypted= ACL condition.
92 JH/17 OpenSSL: the default openssl_options now disables ssl_v3.
94 JH/18 GnuTLS: fix $tls_out_ocsp under hosts_request_ocsp. Previously the
95 verification result was not updated unless hosts_require_ocsp applied.
97 JH/19 Bug 2398: fix listing of a named-queue. Previously, even with the option
98 queue_list_requires_admin set to false, non-admin users were denied the
101 JH/20 Bug 2389: fix server advertising of usable certificates, under GnuTLS in
102 directory-of-certs mode. Previously they were advertised despite the
105 JH/21 The smtp transport option "hosts_noproxy_tls" is now unset by default.
106 A single TCP connection by a client will now hold a TLS connection open
107 for multiple message deliveries, by default. Previoud the default was to
110 JH/22 The smtp transport option "hosts_try_dane" now enables all hosts by
111 default. If built with the facility, DANE will be used. The facility
112 SUPPORT_DANE is now enabled in the prototype build Makefile "EDITME".
114 JH/23 The build default is now for TLS to be included; the SUPPORT_TLS define
115 is replaced with DISABLE_TLS. Either USE_GNUTLS or (the new) USE_OPENSSL
116 must be defined and you must still, unless you define DISABLE_TLS, manage
117 the the include-dir and library-file requirements that go with that
118 choice. Non-TLS builds are still supported.
120 JH/24 Fix duplicated logging of peer name/address, on a transport connection-
123 JH/25 The smtp transport option "hosts_try_fastopen" now enables all hosts by
124 default. If the platform supports and has the facility enabled, it will
125 be requested on all coneections.
127 JH/26 The PIPE_CONNECT facility is promoted from experimental status and is now
128 controlled by the build-time option SUPPORT_PIPE_CONNECT.
130 PP/01 Unbreak heimdal_gssapi, broken in 4.92.
132 JH/27 Bug 2404: Use the main-section configuration option "dsn_from" for
133 success-DSN messages. Previously the From: header was always the default
134 one for these; the option was ignored.
136 JH/28 Fix the timeout on smtp response to apply to the whole response.
137 Previously it was reset for every read, so a teergrubing peer sending
138 single bytes within the time limit could extend the connection for a
139 long time. Credit to Qualsys Security Advisory Team for the discovery.
141 JH/29 Fix DSN Final-Recipient: field. Previously it was the post-routing
142 delivery address, which leaked information of the results of local
143 forwarding. Change to the original envelope recipient address, per
146 JH/30 Bug 2411: Fix DSN generation when RFC 3461 failure notification is
147 requested. Previously not bounce was generated and a log entry of
148 error ignored was made.
150 JH/31 Avoid re-expansion in ${sort } expansion. (CVE-2019-13917)
152 JH/32 Introduce a general tainting mechanism for values read from the input
153 channel, and values derived from them. Refuse to expand any tainted
154 values, to catch one form of exploit.
156 JH/33 Bug 2413: Fix dkim_strict option. Previously the expansion result
157 was unused and the unexpanded text used for the test. Found and
158 fixed by Ruben Jenster.
160 JH/34 Fix crash after TLS shutdown. When the TCP/SMTP channel was left open,
161 an attempt to use a TLS library read routine dereffed a nul pointer,
164 JH/35 Bug 2409: filter out-of-spec chars from callout response before using
165 them in our smtp response.
167 JH/36 Have the general router option retry_use_local_part default to true when
168 any of the restrictive preconditions are set (to anything). Previously it
169 was only for check_local user. The change removes one item of manual
170 configuration which is required for proper retries when a remote router
171 handles a subset of addresses for a domain.
173 JH/37 Appendfile: when evaluating quota use (non-quota_size_regex) take the file
174 link count into consideration.
176 HS/04 Fix handling of very log lines in -H files. If a -<key> <value> line
177 caused the extension of big_buffer, the following lines were ignored.
179 JH/38 Bug 1395: Teach the DNS negative-cache about TTL value from the SOA in
180 accordance with RFC 2308. Previously there was no expiry, so a longlived
181 receive process (eg. due to ACL delays) versus a short SOA value could
184 HS/05 Handle trailing backslash gracefully. (CVE-2019-15846)
186 JH/39 Promote DMARC support to mainline.
188 JH/40 Bug 2452: Add a References: header to DSNs.
194 JH/01 Remove code calling the customisable local_scan function, unless a new
195 definition "HAVE_LOCAL_SCAN=yes" is present in the Local/Makefile.
197 JH/02 Bug 1007: Avoid doing logging from signal-handlers, as that can result in
198 non-signal-safe functions being used.
200 JH/03 Bug 2269: When presented with a received message having a stupidly large
201 number of DKIM-Signature headers, disable DKIM verification to avoid
202 a resource-consumption attack. The limit is set at twenty.
204 JH/04 Add variables $arc_domains, $arc_oldest_pass for ARC verify. Fix the
205 report of oldest_pass in ${authres } in consequence, and separate out
206 some descriptions of reasons for verification fail.
208 JH/05 Bug 2273: Cutthrough delivery left a window where the received messsage
209 files in the spool were present and unlocked. A queue-runner could spot
210 them, resulting in a duplicate delivery. Fix that by doing the unlock
211 after the unlink. Investigation by Tim Stewart. Take the opportunity to
212 add more error-checking on spoolfile handling while that code is being
215 PP/01 Refuse to open a spool data file (*-D) if it's a symlink.
216 No known attacks, no CVE, this is defensive hardening.
218 JH/06 Bug 2275: The MIME ACL unlocked the received message files early, and
219 a queue-runner could start a delivery while other operations were ongoing.
220 Cutthrough delivery was a common victim, resulting in duplicate delivery.
221 Found and investigated by Tim Stewart. Fix by using the open message data
222 file handle rather than opening another, and not locally closing it (which
223 releases a lock) for that case, while creating the temporary .eml format
224 file for the MIME ACL. Also applies to "regex" and "spam" ACL conditions.
226 JH/07 Bug 177: Make a random-recipient callout success visible in ACL, by setting
227 $sender_verify_failure/$recipient_verify_failure to "random".
229 JH/08 When generating a selfsigned cert, use serial number 1 since zero is not
232 JH/09 Bug 2274: Fix logging of cmdline args when starting in an unlinked cwd.
233 Previously this would segfault.
235 JH/10 Fix ARC signing for case when DKIM signing failed. Previously this would
238 JH/11 Bug 2264: Exim now only follows CNAME chains one step by default. We'd
239 like zero, since the resolver should be doing this for us, But we need one
240 as a CNAME but no MX presence gets the CNAME returned; we need to check
241 that doesn't point to an MX to declare it "no MX returned" rather than
242 "error, loop". A new main option is added so the older capability of
243 following some limited number of chain links is maintained.
245 JH/12 Add client-ip info to non-pass iprev ${authres } lines.
247 JH/13 For receent Openssl versions (1.1 onward) use modern generic protocol
248 methods. These should support TLS 1.3; they arrived with TLS 1.3 and the
249 now-deprecated earlier definitions used only specified the range up to TLS
250 1.2 (in the older-version library docs).
252 JH/14 Bug 2284: Fix DKIM signing for body lines starting with a pair of dots.
254 JH/15 Rework TLS client-side context management. Stop using a global, and
255 explicitly pass a context around. This enables future use of TLS for
256 connections to service-daemons (eg. malware scanning) while a client smtp
257 connection is using TLS; with cutthrough connections this is quite likely.
259 JH/16 Fix ARC verification to do AS checks in reverse order.
261 JH/17 Support a "tls" option on the ${readsocket } expansion item.
263 JH/18 Bug 2287: Fix the protocol name (eg utf8esmtp) for multiple messages
264 using the SMTPUTF8 option on their MAIL FROM commands, in one connection.
265 Previously the "utf8" would be re-prepended for every additional message.
267 JH/19 Reject MAIL FROM commands with SMTPUTF8 when the facility was not advertised.
268 Previously thery were accepted, resulting in issues when attempting to
269 forward messages to a non-supporting MTA.
271 PP/02 Let -n work with printing macros too, not just options.
273 JH/20 Bug 2296: Fix cutthrough for >1 address redirection. Previously only
274 one parent address was copied, and bogus data was used at delivery-logging
275 time. Either a crash (after delivery) or bogus log data could result.
276 Discovery and analysis by Tim Stewart.
278 PP/03 Make ${utf8clean:} expansion operator detect incomplete final character.
279 Previously if the string ended mid-character, we did not insert the
280 promised '?' replacement.
282 PP/04 Documentation: current string operators work on bytes, not codepoints.
284 JH/21 Change as many as possible of the global flags into one-bit bitfields; these
285 should pack well giving a smaller memory footprint so better caching and
286 therefore performance. Group the declarations where this can't be done so
287 that the byte-sized flag variables are not interspersed among pointer
288 variables, giving a better chance of good packing by the compiler.
290 JH/22 Bug 1896: Fix the envelope from for DMARC forensic reports to be possibly
291 non-null, to avoid issues with sites running BATV. Previously reports were
292 sent with an empty envelope sender so looked like bounces.
294 JH/23 Bug 2318: Fix the noerror command within filters. It wasn't working.
295 The ignore_error flag wasn't being returned from the filter subprocess so
296 was not set for later routers. Investigation and fix by Matthias Kurz.
298 JH/24 Bug 2310: Raise a msg:fail:internal event for each undelivered recipient,
299 and a msg:complete for the whole, when a message is manually removed using
300 -Mrm. Developement by Matthias Kurz, hacked on by JH.
302 JH/25 Avoid fixed-size buffers for pathnames in DB access. This required using
303 a "Gnu special" function, asprintf() in the DB utility binary builds; I
304 hope that is portable enough.
306 JH/26 Bug 2311: Fix DANE-TA verification under GnuTLS. Previously it was also
307 requiring a known-CA anchor certificate; make it now rely entirely on the
308 TLSA as an anchor. Checking the name on the leaf cert against the name
309 on the A-record for the host is still done for TA (but not for EE mode).
311 JH/27 Fix logging of proxy address. Previously, a pointless "PRX=[]:0" would be
312 included in delivery lines for non-proxied connections, when compiled with
313 SUPPORT_SOCKS and running with proxy logging enabled.
315 JH/28 Bug 2314: Fire msg:fail:delivery event even when error is being ignored.
316 Developement by Matthias Kurz, tweaked by JH. While in that bit of code,
317 move the existing event to fire before the normal logging of message
318 failure so that custom logging is bracketed by normal logging.
320 JH/29 Bug 2322: A "fail" command in a non-system filter (file) now fires the
321 msg:fail:internal event. Developement by Matthias Kurz.
323 JH/30 Bug 2329: Increase buffer size used for dns lookup from 2k, which was
324 far too small for todays use of crypto signatures stored there. Go all
325 the way to the max DNS message size of 64kB, even though this might be
326 overmuch for IOT constrained device use.
328 JH/31 Fix a bad use of a copy function, which could be used to pointlessly
329 copy a string over itself. The library routine is documented as not
330 supporting overlapping copies, and on MacOS it actually raised a SIGABRT.
332 JH/32 For main options check_spool_space and check_inode_space, where the
333 platform supports 64b integers, support more than the previous 2^31 kB
334 (i.e. more than 2 TB). Accept E, P and T multipliers in addition to
335 the previous G, M, k.
337 JH/33 Bug 2338: Fix the cyrus-sasl authenticator to fill in the
338 $authenticated_fail_id variable on authentication failure. Previously
341 JH/34 Increase RSA keysize of autogen selfsign cert from 1024 to 2048. RHEL 8.0
342 OpenSSL didn't want to use such a weak key. Do for GnuTLS also, and for
343 more-modern GnuTLS move from GNUTLS_SEC_PARAM_LOW to
344 GNUTLS_SEC_PARAM_MEDIUM.
346 JH/35 OpenSSL: fail the handshake when SNI processing hits a problem, server
347 side. Previously we would continue as if no SNI had been received.
349 JH/36 Harden the handling of string-lists. When a list consisted of a sole
350 "<" character, which should be a list-separator specification, we walked
351 off past the nul-terimation.
353 JH/37 Bug 2341: Send "message delayed" warning MDNs (restricted to external
354 causes) even when the retry time is not yet met. Previously they were
355 not, meaning that when (say) an account was over-quota and temp-rejecting,
356 and multiple senders' messages were queued, only one sender would get
357 notified on each configured delay_warning cycle.
359 JH/38 Bug 2351: Log failures to extract envelope addresses from message headers.
361 JH/39 OpenSSL: clear the error stack after an SSL_accept(). With anon-auth
362 cipher-suites, an error can be left on the stack even for a succeeding
363 accept; this results in impossible error messages when a later operation
366 AM/01 Bug 2359: GnuTLS: repeat lowlevel read and write operations while they
367 return error codes indicating retry. Under TLS1.3 this becomes required.
369 JH/40 Fix the feature-cache refresh for EXPERIMENTAL_PIPE_CONNECT. Previously
370 it only wrote the new authenticators, resulting in a lack of tracking of
371 peer changes of ESMTP extensions until the next cache flush.
373 JH/41 Fix the loop reading a message header line to check for integer overflow,
374 and more-often against header_maxsize. Previously a crafted message could
375 induce a crash of the recive process; now the message is cleanly rejected.
377 JH/42 Bug 2366: Fix the behaviour of the dkim_verify_signers option. It had
378 been totally disabled for all of 4.91. Discovery and fix by "Mad Alex".
384 GF/01 DEFER rather than ERROR on redis cluster MOVED response.
385 When redis_servers is set to a list of > 1 element, and the Redis servers
386 in that list are in cluster configuration, convert the REDIS_REPLY_ERROR
387 case of MOVED into a DEFER case instead, thus moving the query onto the
388 next server in the list. For a cluster of N elements, all N servers must
389 be defined in redis_servers.
391 GF/02 Catch and remove uninitialized value warning in exiqsumm
392 Check for existence of @ARGV before looking at $ARGV[0]
394 JH/01 Replace the store_release() internal interface with store_newblock(),
395 which internalises the check required to safely use the old one, plus
396 the allocate and data copy operations duplicated in both (!) of the
397 extant use locations.
399 JH/02 Disallow '/' characters in queue names specified for the "queue=" ACL
400 modifier. This matches the restriction on the commandline.
402 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
403 Previously only the last row was returned.
405 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
406 we assumed that tags in the header were well-formed, and parsed the
407 element content after inspecting only the first char of the tag.
408 Assumptions at that stage could crash the receive process on malformed
411 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
412 While running the DKIM ACL we operate on the Permanent memory pool so that
413 variables created with "set" persist to the DATA ACL. Also (at any time)
414 DNS lookups that fail create cache records using the Permanent pool. But
415 expansions release any allocations made on the current pool - so a dnsdb
416 lookup expansion done in the DKIM ACL releases the memory used for the
417 DNS negative-cache, and bad things result. Solution is to switch to the
418 Main pool for expansions.
419 While we're in that code, add checks on the DNS cache during store_reset,
420 active in the testsuite.
421 Problem spotted, and debugging aided, by Wolfgang Breyha.
423 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
424 When none of the hosts presented to a transport match an already-open
425 connection, close it and proceed with the list. Previously we would
426 queue the message. Spotted by Lena with Yahoo, probably involving
429 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
430 Previously a spurious "250 OK id=" response was appended to the proper
433 JH/08 The "support for" informational output now, which built with Content
434 Scanning support, has a line for the malware scanner interfaces compiled
435 in. Interface can be individually included or not at build time.
437 JH/09 The "aveserver", "kavdaemon" and "mksd" interfaces are now not included
438 by the template makefile "src/EDITME". The "STREAM" support for an older
439 ClamAV interface method is removed.
441 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
442 rows affected is given instead).
444 JH/11 The runtime Berkeley DB library version is now additionally output by
445 "exim -d -bV". Previously only the compile-time version was shown.
447 JH/12 Bug 2230: Fix cutthrough routing for nonfirst messages in an initiating
448 SMTP connection. Previously, when one had more receipients than the
449 first, an abortive onward connection was made. Move to full support for
450 multiple onward connections in sequence, handling cutthrough connection
451 for all multi-message initiating connections.
453 JH/13 Bug 2229: Fix cutthrough routing for nonstandard port numbers defined by
454 routers. Previously, a multi-recipient message would fail to match the
455 onward-connection opened for the first recipient, and cause its closure.
457 JH/14 Bug 2174: A timeout on connect for a callout was also erroneously seen as
458 a timeout on read on a GnuTLS initiating connection, resulting in the
459 initiating connection being dropped. This mattered most when the callout
460 was marked defer_ok. Fix to keep the two timeout-detection methods
463 JH/15 Relax results from ACL control request to enable cutthrough, in
464 unsupported situations, from error to silently (except under debug)
465 ignoring. This covers use with PRDR, frozen messages, queue-only and
468 HS/01 Fix Buffer overflow in base64d() (CVE-2018-6789)
470 JH/16 Fix bug in DKIM verify: a buffer overflow could corrupt the malloc
471 metadata, resulting in a crash in free().
473 PP/01 Fix broken Heimdal GSSAPI authenticator integration.
474 Broken in f2ed27cf5, missing an equals sign for specified-initialisers.
475 Broken also in d185889f4, with init system revamp.
477 JH/17 Bug 2113: Fix conversation closedown with the Avast malware scanner.
478 Previously we abruptly closed the connection after reading a malware-
479 found indication; now we go on to read the "scan ok" response line,
482 JH/18 Bug 2239: Enforce non-usability of control=utf8_downconvert in the mail
483 ACL. Previously, a crash would result.
485 JH/19 Speed up macro lookups during configuration file read, by skipping non-
486 macro text after a replacement (previously it was only once per line) and
487 by skipping builtin macros when searching for an uppercase lead character.
489 JH/20 DANE support moved from Experimental to mainline. The Makefile control
490 for the build is renamed.
492 JH/21 Fix memory leak during multi-message connections using STARTTLS. A buffer
493 was allocated for every new TLS startup, meaning one per message. Fix
494 by only allocating once (OpenSSL) or freeing on TLS-close (GnuTLS).
496 JH/22 Bug 2236: When a DKIM verification result is overridden by ACL, DMARC
497 reported the original. Fix to report (as far as possible) the ACL
498 result replacing the original.
500 JH/23 Fix memory leak during multi-message connections using STARTTLS under
501 OpenSSL. Certificate information is loaded for every new TLS startup,
502 and the resources needed to be freed.
504 JH/24 Bug 2242: Fix exim_dbmbuild to permit directoryless filenames.
506 JH/25 Fix utf8_downconvert propagation through a redirect router. Previously it
509 JH/26 Bug 2253: For logging delivery lines under PRDR, append the overall
510 DATA response info to the (existing) per-recipient response info for
511 the "C=" log element. It can have useful tracking info from the
512 destination system. Patch from Simon Arlott.
514 JH/27 Bug 2251: Fix ldap lookups that return a single attribute having zero-
515 length value. Previously this would segfault.
517 HS/02 Support Avast multiline protoocol, this allows passing flags to
518 newer versions of the scanner.
520 JH/28 Ensure that variables possibly set during message acceptance are marked
521 dead before release of memory in the daemon loop. This stops complaints
522 about them when the debug_store option is enabled. Discovered specifically
523 for sender_rate_period, but applies to a whole set of variables.
524 Do the same for the queue-runner and queue-list loops, for variables set
525 from spool message files. Do the same for the SMTP per-message loop, for
526 certain variables indirectly set in ACL operations.
528 JH/29 Bug 2250: Fix a longstanding bug in heavily-pipelined SMTP input (such
529 as a multi-recipient message from a mailinglist manager). The coding had
530 an arbitrary cutoff number of characters while checking for more input;
531 enforced by writing a NUL into the buffer. This corrupted long / fast
532 input. The problem was exposed more widely when more pipelineing of SMTP
533 responses was introduced, and one Exim system was feeding another.
534 The symptom is log complaints of SMTP syntax error (NUL chars) on the
535 receiving system, and refused recipients seen by the sending system
536 (propating to people being dropped from mailing lists).
537 Discovered and pinpointed by David Carter.
539 JH/30 The (EXPERIMENTAL_DMARC) variable $dmarc_ar_header is withdrawn, being
540 replaced by the ${authresults } expansion.
542 JH/31 Bug 2257: Fix pipe transport to not use a socket-only syscall.
544 HS/03 Set a handler for SIGTERM and call exit(3) if running as PID 1. This
545 allows proper process termination in container environments.
547 JH/32 Bug 2258: Fix spool_wireformat in combination with LMTP transport.
548 Previously the "final dot" had a newline after it; ensure it is CR,LF.
550 JH/33 SPF: remove support for the "spf" ACL condition outcome values "err_temp"
551 and "err_perm", deprecated since 4.83 when the RFC-defined words
552 "temperror" and "permerror" were introduced.
554 JH/34 Re-introduce enforcement of no cutthrough delivery on transports having
555 transport-filters or DKIM-signing. The restriction was lost in the
556 consolidation of verify-callout and delivery SMTP handling.
557 Extend the restriction to also cover ARC-signing.
559 JH/35 Cutthrough: for a final-dot response timeout (and nonunderstood responses)
560 in defer=pass mode supply a 450 to the initiator. Previously the message
563 PP/02 DANE: add dane_require_tls_ciphers SMTP Transport option; if unset,
564 tls_require_ciphers is used as before.
566 HS/03 Malware Avast: Better match the Avast multiline protocol. Add
567 "pass_unscanned". Only tmpfails from the scanner are written to
568 the paniclog, as they may require admin intervention (permission
569 denied, license issues). Other scanner errors (like decompression
570 bombs) do not cause a paniclog entry.
572 JH/36 Fix reinitialisation of DKIM logging variable between messages.
573 Previously it was possible to log spurious information in receive log
576 JH/37 Bug 2255: Revert the disable of the OpenSSL session caching. This
577 triggered odd behaviour from Outlook Express clients.
579 PP/03 Add util/renew-opendmarc-tlds.sh script for safe renewal of public
582 JH/38 DKIM: accept Ed25519 pubkeys in SubjectPublicKeyInfo-wrapped form,
583 since the IETF WG has not yet settled on that versus the original
584 "bare" representation.
586 JH/39 Fix syslog logging for syslog_timestamp=no and log_selector +millisec.
587 Previously the millisecond value corrupted the output.
588 Fix also for syslog_pid=no and log_selector +pid, for which the pid
589 corrupted the output.
595 JH/01 Rework error string handling in TLS interface so that the caller in
596 more cases is responsible for logging. This permits library-sourced
597 string to be attached to addresses during delivery, and collapses
598 pairs of long lines into single ones.
600 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
601 during configuration. Wildcards are allowed and expanded.
603 JH/02 Rework error string handling in DKIM to pass more info back to callers.
604 This permits better logging.
606 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
607 do not close it down and have the child transport start it up again on
608 the passed-on TCP connection. Instead, proxy the child (and any
609 subsequent ones) for TLS via a unix-domain socket channel. Logging is
610 affected: the continued delivery log lines do not have any DNSSEC, TLS
611 Certificate or OCSP information. TLS cipher information is still logged.
613 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
614 identical IP addresses on different listening ports. Will also affect
617 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
618 add noisy ifdef guards to special-case this sillyness.
619 Patch from Bernd Kuhls.
621 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
622 than 255 are no longer allowed.
624 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
625 Disable the session-cache too, which might reduce our load. Since we
626 currrectly use a new context for every connection, both as server and
627 client, there is no benefit for these.
628 GnuTLS appears to not support tickets server-side by default (we don't
629 call gnutls_session_ticket_enable_server()) but client side is enabled
630 by default on recent versions (3.1.3 +) unless the PFS priority string
633 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
634 <https://reproducible-builds.org/specs/source-date-epoch/>.
636 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
637 the check for any unsuccessful recipients did not notice the limit, and
638 erroneously found still-pending ones.
640 JH/08 Pipeline CHUNKING command and data together, on kernels that support
641 MSG_MORE. Only in-clear (not on TLS connections).
643 JH/09 Avoid using a temporary file during transport using dkim. Unless a
644 transport-filter is involved we can buffer the headers in memory for
645 creating the signature, and read the spool data file once for the
646 signature and again for transmission.
648 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
649 4.77 as the kernel support then wasn't solid, having issues in 64bit
650 mode. Now, it's been long enough. Add support for FreeBSD also.
652 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
653 case where the routing stage had gathered several addresses to send to
654 a host before calling the transport for the first, we previously failed
655 to close down TLS in the old transport process before passing the TCP
656 connection to the new process. The new one sent a STARTTLS command
657 which naturally failed, giving a failed delivery and bloating the retry
658 database. Investigation and fix prototype from Wolfgang Breyha.
660 JH/12 Fix check on SMTP command input synchronisation. Previously there were
661 false-negatives in the check that the sender had not preempted a response
662 or prompt from Exim (running as a server), due to that code's lack of
663 awareness of the SMTP input buffering.
665 PP/04 Add commandline_checks_require_admin option.
666 Exim drops privileges sanely, various checks such as -be aren't a
667 security problem, as long as you trust local users with access to their
668 own account. When invoked by services which pass untrusted data to
669 Exim, this might be an issue. Set this option in main configuration
670 AND make fixes to the calling application, such as using `--` to stop
673 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
674 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
675 responses to those, into a single TLS record each way (this usually means
676 a single packet). As a side issue, smtp_enforce_sync now works on TLS
679 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
680 affects you only if you're dancing at the edge of the param size limits.
681 If you are, and this message makes sense to you, then: raise the
682 configured limit or use OpenSSL 1.1. Nothing we can do for older
685 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
686 cmdline element to get the documented default one. Previously it was
689 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
690 are unsuable for non-message-specific reasons - eg. connection timeout,
691 banner-time rejection.
693 JH/16 Fix logging of delivery remote port, when specified by router, under
696 PP/06 Repair manualroute's ability to take options in any order, even if one
697 is the name of a transport.
700 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
702 JH/17 Change the list-building routines interface to use the expanding-string
703 triplet model, for better allocation and copying behaviour.
705 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
706 Previously it was constructed the first time a possibly-matching string
707 was met in the configuration file input during startup; now it is done
710 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
711 compatible one, to avoid the (poorly documented) possibility of a config
712 file in the working directory redirecting the DB files, possibly correpting
713 some existing file. CVE-2017-10140 assigned for BDB.
715 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
716 cache-hot. Previously, although the result was properly cached, the
717 initial verify call returned a defer.
719 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
720 the main verify for receipient in uncached-mode.
722 JH/22 Retire historical build files to an "unsupported" subdir. These are
723 defined as "ones for which we have no current evidence of testing".
725 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
726 if present. Previously it was ignored.
728 JH/24 Start using specified-initialisers in C structure init coding. This is
729 a C99 feature (it's 2017, so now considered safe).
731 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
732 if was a fixed-sized field and bitmask ops via macros; it is now more
735 PP/07 GitHub PR 56: Apply MariaDB build fix.
736 Patch provided by Jaroslav Škarvada.
738 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
739 during Coverity cleanups [4.87 JH/47]
740 Diagnosis and fix provided by Michael Fischer v. Mollard.
742 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
743 the right size to place the terminating semicolon on its own folded
744 line, the header hash was calculated to an incorrect value thanks to
745 the (relaxed) space the fold became.
747 HS/02 Fix Bug 2130: large writes from the transport subprocess were chunked
748 and confused the parent.
750 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
751 which could crash as a result. This could lead to undeliverable messages.
753 JH/28 Logging: "next input sent too soon" now shows where input was truncated
756 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
757 matters on fast-turnover and PID-randomising systems, which were getting
758 out-of-order delivery.
760 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
761 a possibly-overlapping copy. The symptom was that "Remote host closed
762 connection in response to HELO" was logged instead of the actual 4xx
765 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
766 Previously only that bufferd was discarded, resulting in SYMTP command
769 JH/32 DKIM: when a message has multiple signatures matching an identity given
770 in dkim_verify_signers, run the dkim acl once for each. Previously only
771 one run was done. Bug 2189.
773 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
774 "panic the current process" to "deliberately defer". The panic log is
775 still written with the problem list name; the mail and reject logs now
776 get a temp-reject line for the message that was being handled, saying
777 something like "domains check lookup or other defer". The SMTP 451
778 message is still "Temporary local problem".
780 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
781 A crafted sequence of BDAT commands could result in in-use memory beeing
782 freed. CVE-2017-16943.
784 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
785 from SMTP input. Previously it was always done; now only done for DATA
786 and not BDAT commands. CVE-2017-16944.
788 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
789 to the message (such as an overlong header line). Previously this was
790 not done and we did not exit BDAT mode. Followon from the previous item
791 though a different problem.
797 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
798 than -2003 did; needs libidn2 in addition to libidn.
800 JH/02 The path option on a pipe transport is now expanded before use.
802 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
803 Patch provided by "Björn", documentation fix added too.
805 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
806 missing a wire-to-host endian conversion.
808 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
809 close after a BDAT command line could be taken as a following command,
810 giving a synch failure. Fix by only checking for synch immediately
811 before acknowledging the chunk.
813 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
814 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
815 macro. Patches provided by Josh Soref.
817 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
818 Previously we did not; the RFC seems ambiguous and VRFY is not listed
819 by IANA as a service extension. However, John Klensin suggests that we
822 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
823 the dkim code may be unix-mode line endings rather than smtp wire-format
824 CRLF, so prepend a CR to any bare LF.
826 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
827 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
829 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
830 the first were themselves being wrongly included in the feed into dkim
831 processing; with most chunk sizes in use this resulted in an incorrect
832 body hash calculated value.
834 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
835 DKIM signature block, for verification. Although advised against by
836 standards it is specifically not ruled illegal.
838 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
840 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
841 missing a body hash (the bh= tag).
843 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
844 It seems that HAProxy sends the Proxy Protocol information in clear and
845 only then does a TLS startup, so do the same.
847 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
848 TCP connections (such as for Spamd) unless the daemon successfully set
849 Fast Open mode on its listening sockets. This fixes breakage seen on
850 too-old kernels or those not configured for Fast Open, at the cost of
851 requiring both directions being enabled for TFO, and TFO never being used
852 by non-daemon-related Exim processes.
854 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
855 endings, at least on the first header line. Try to canonify any that get
856 past that check, despite the cost.
858 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
859 now limited to an arbitrary five deep, while parsing addresses with the
860 strip_excess_angle_brackets option enabled.
862 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
863 instead leave the unprompted TLS handshake in socket buffer for the
864 TLS library to consume.
866 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
868 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
870 JH/16 Drop variables when they go out of scope. Memory management drops a whole
871 region in one operation, for speed, and this leaves assigned pointers
872 dangling. Add checks run only under the testsuite which checks all
873 variables at a store-reset and panics on a dangling pointer; add code
874 explicitly nulling out all the variables discovered. Fixes one known
875 bug: a transport crash, where a dangling pointer for $sending_ip_address
876 originally assigned in a verify callout, is re-used.
878 PP/06 Drop '.' from @INC in various Perl scripts.
880 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
882 PP/08 Reduce a number of compilation warnings under clang; building with
883 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
884 should be warning-free.
886 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
888 HS/01 Fix portability problems introduced by PP/08 for platforms where
889 realloc(NULL) is not equivalent to malloc() [SunOS et al].
891 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
892 chunk. This allows us to accept broken chunked messages. We need a more
893 general solution here.
895 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
896 already-broken messages in the queue.
898 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
900 JH/19 Fix reference counting bug in routing-generated-address tracking.
906 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
907 supports it and a size is available (ie. the sending peer gave us one).
909 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
910 years of being deprecated). The replacements are the ACLs
911 acl_smtp_mime and acl_not_smtp_mime.
913 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
914 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
915 or even an in-clear connection were permitted. Now, if the host lookup
916 was dnssec and dane was requested then the host is only used if the
917 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
918 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
919 if one fails this test.
920 This means that a poorly-configured remote DNS will make it incommunicado;
921 but it protects against a DNS-interception attack on it.
923 JH/04 Bug 1810: make continued-use of an open smtp transport connection
924 non-noisy when a race steals the message being considered.
926 JH/05 If main configuration option tls_certificate is unset, generate a
927 self-signed certificate for inbound TLS connections.
929 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
930 in rewrites and routers.
932 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
933 and logged a warning sing 4.83; now they are a configuration file error.
935 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
936 (lacking @domain). Apply the same qualification processing as RCPT.
938 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
940 JH/10 Support ${sha256:} applied to a string (as well as the previous
943 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
944 a cutthrough deliver is pending, as we always want to make a connection.
945 This also avoids re-routing the message when later placing the cutthrough
946 connection after a verify cache hit.
947 Do not update it with the verify result either.
949 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
950 when routing results in more than one destination address.
952 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
953 signing (which inhibits the cutthrough capability). Previously only
954 the presence of an option was tested; now an expansion evaluating as
955 empty is permissible (obviously it should depend only on data available
956 when the cutthrough connection is made).
958 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
959 the relevant preceding SMTP command did not note the pipelining mode.
961 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
962 Previously they were not counted.
964 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
965 as one having no matching records. Previously we deferred the message
966 that needed the lookup.
968 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
969 distinguished as "(=".
971 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
972 for missing MX records. Previously it only worked for missing A records.
974 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
976 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
977 after the data-go-ahead and data-ack. Patch from Jason Betts.
979 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
980 even for a "none" policy. Patch from Tony Meyer.
982 JH/22 Fix continued use of a connection for further deliveries. If a port was
983 specified by a router, it must also match for the delivery to be
986 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
987 When one of the recipients of a message was unsuitable for the connection
988 (has no matching addresses), we lost track of needing to mark it
989 deferred. As a result mail would be lost.
991 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
993 JH/25 Decoding ACL controls is now done using a binary search; the source code
994 takes up less space and should be simpler to maintain. Merge the ACL
995 condition decode tables also, with similar effect.
997 JH/26 Fix problem with one_time used on a redirect router which returned the
998 parent address unchanged. A retry would see the parent address marked as
999 delivered, so not attempt the (identical) child. As a result mail would
1002 JH/27 Fix a possible security hole, wherein a process operating with the Exim
1003 UID can gain a root shell. Credit to http://www.halfdog.net/ for
1004 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
1007 JH/28 Enable {spool,log} filesystem space and inode checks as default.
1008 Main config options check_{log,spool}_{inodes,space} are now
1009 100 inodes, 10MB unless set otherwise in the configuration.
1011 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
1012 Previously it only applied to the main-section connection policy
1015 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
1017 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
1018 by me. Added RFC7919 DH primes as an alternative.
1020 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
1021 are not in the system include path.
1023 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
1024 GnuTLS, when a session startup failed (eg because the client disconnected)
1025 Exim did stdio operations after fclose. This was exposed by a recent
1026 change which nulled out the file handle after the fclose.
1028 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
1029 signed directly by the cert-signing cert, rather than an intermediate
1030 OCSP-signing cert. This is the model used by LetsEncrypt.
1032 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
1034 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
1035 an incoming connection.
1037 HS/02 Bug 1802: Do not half-close the connection after sending a request
1040 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
1041 fallback to "prime256v1".
1043 JH/34 SECURITY: Use proper copy of DATA command in error message.
1044 Could leak key material. Remotely exploitable. CVE-2016-9963.
1050 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
1051 and 3.4.4 - once the server is enabled to respond to an OCSP request
1052 it does even when not requested, resulting in a stapling non-aware
1053 client dropping the TLS connection.
1055 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
1056 support variable-length bit vectors. No functional change.
1058 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
1059 The I= interface field on outgoing lines is now after the H= remote
1060 host field, same as incoming lines. There is a separate
1061 outgoing_interface log selector which allows you to disable the
1064 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
1065 If not running log_selector +smtp_connection the mainlog would be held
1066 open indefinitely after a "too many connections" event, including to a
1067 deleted file after a log rotate. Leave the per net connection logging
1068 leaving it open for efficiency as that will be quickly detected by the
1069 check on the next write.
1071 HS/01 Bug 1671: Fix post transport crash.
1072 Processing the wait-<transport> messages could crash the delivery
1073 process if the message IDs didn't exist for some reason. When
1074 using 'split_spool_directory=yes' the construction of the spool
1075 file name failed already, exposing the same netto behaviour.
1077 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
1078 mime_regex ACL conditions.
1080 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
1081 to DSN fail messages (bounces): remote IP, remote greeting, remote response
1082 to HELO, local diagnostic string.
1084 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
1085 log line to debug. Even when configured with a tls authenticator many
1086 client connections are expected to not authenticate in this way, so
1087 an authenticate fail is not an error.
1089 HS/02 Add the Exim version string to the process info. This way exiwhat
1090 gives some more detail about the running daemon.
1092 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
1093 matter for fast-change records such as DNSBLs.
1095 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
1096 retry record, even if constant. There may be multiple transports with
1097 different interface settings and the retry behaviour needs to be kept
1100 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
1102 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
1104 JH/10 Bug 840: fix log_defer_output option of pipe transport
1106 JH/11 Bug 830: use same host for all RCPTS of a message, even under
1107 hosts_randomize. This matters a lot when combined with mua_wrapper.
1109 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
1110 ${quote_pgsql:<string>} operator.
1112 JH/13 Bug 1708: avoid misaligned access in cached lookup.
1114 JH/14 Change header file name for freeradius-client. Relevant if compiling
1115 with Radius support; from the Gentoo tree and checked under Fedora.
1117 JH/15 Bug 1712: Introduce $prdr_requested flag variable
1119 JH/16 Bug 1714: Permit an empty string as expansion result for transport
1120 option transport_filter, meaning no filtering.
1122 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
1124 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
1125 defaults to "*" (all hosts). The variable is now available when not built
1126 with TLS, default unset, mainly to enable keeping the testsuite sane.
1127 If a server certificate is not supplied (via tls_certificate) an error is
1128 logged, and clients will find TLS connections fail on startup. Presumably
1129 they will retry in-clear.
1130 Packagers of Exim are strongly encouraged to create a server certificate
1131 at installation time.
1133 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
1134 with the $config_file variable.
1136 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
1137 in transport context, after the attempt, and per-recipient. The latter type
1138 is per host attempted. The event data is the error message, and the errno
1139 information encodes the lookup type (A vs. MX) used for the (first) host,
1140 and the trailing two digits of the smtp 4xx response.
1142 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
1143 to write to mainlog (or rejectlog, paniclog) in the window between file
1144 creation and permissions/ownership being changed. Particularly affects
1145 installations where exicyclog is run as root, rather than exim user;
1146 result is that the running daemon panics and dies.
1148 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
1150 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
1151 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
1152 "pri" and "weight". Note that the previous implicit priority given by the
1153 list order is no longer honoured.
1155 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
1156 for DKIM processing.
1158 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
1159 by defining SUPPORT_SOCKS.
1161 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
1162 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
1163 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
1164 variables are renamed to proxy_{local,external}_{address,port}.
1166 JH/27 Move Internationalisation support from Experimental to mainline, enabled
1167 for a build by defining SUPPORT_I18N
1169 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
1170 of the query string, and make ${quote_redis:} do that quoting.
1172 JH/29 Move Events support from Experimental to mainline, enabled by default
1173 and removable for a build by defining DISABLE_EVENT.
1175 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
1177 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
1178 cached by the daemon.
1180 JH/32 Move Redis support from Experimental to mainline, enabled for a build
1181 by defining LOOKUP_REDIS. The libhiredis library is required.
1183 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
1184 keys are given for lookup.
1186 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
1187 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
1188 only supported when built with TLS support. The PolarSSL SHA routines
1189 are still used when the TLS library is too old for convenient support.
1191 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
1192 openssl_options), for security. OpenSSL forces this from version 1.1.0
1193 server-side so match that on older versions.
1195 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
1196 allocation for $value could be released as the expansion processing
1197 concluded, but leaving the global pointer active for it.
1199 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
1200 and to use the domains and local_parts ACL conditions.
1202 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
1203 incorrectly not doubled on cutthrough transmission, hence seen as a
1204 body-termination at the receiving system - resulting in truncated mails.
1205 Commonly the sender saw a TCP-level error, and retransmitted the message
1206 via the normal store-and-forward channel. This could result in duplicates
1207 received - but deduplicating mailstores were liable to retain only the
1208 initial truncated version.
1210 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
1212 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
1214 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
1215 we're in there, support oversigning also; bug 1309.
1217 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
1219 HS/04 Add support for keep_environment and add_environment options.
1221 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
1222 either intentional arithmetic overflow during PRNG, or testing config-
1225 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
1226 delivery resulted in actual delivery. Cancel cutthrough before DATA
1229 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
1230 rejects a recipient: pass the reject to the originator.
1232 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
1233 Many were false-positives and ignorable, but it's worth fixing the
1236 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
1237 for the new environment-manipulation done at startup. Move the routines
1238 from being local to tls.c to being global via the os.c file.
1240 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
1241 an extract embedded as result-arg for a map, the first arg for extract
1242 is unavailable so we cannot tell if this is a numbered or keyed
1243 extraction. Accept either.
1249 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
1252 JH/02 The smtp transport option "multi_domain" is now expanded.
1254 JH/03 The smtp transport now requests PRDR by default, if the server offers
1257 JH/04 Certificate name checking on server certificates, when exim is a client,
1258 is now done by default. The transport option tls_verify_cert_hostnames
1259 can be used to disable this per-host. The build option
1260 EXPERIMENTAL_CERTNAMES is withdrawn.
1262 JH/05 The value of the tls_verify_certificates smtp transport and main options
1263 default to the word "system" to access the system default CA bundle.
1264 For GnuTLS, only version 3.0.20 or later.
1266 JH/06 Verification of the server certificate for a TLS connection is now tried
1267 (but not required) by default. The verification status is now logged by
1268 default, for both outbound TLS and client-certificate supplying inbound
1271 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
1274 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
1275 Status Notification (bounce) messages are now MIME format per RFC 3464.
1276 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
1277 under the control of the dsn_advertise_hosts option, and routers may
1278 have a dsn_lasthop option.
1280 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
1281 default, modifiable by a malware= option. The list separator for
1282 the options can now be changed in the usual way. Bug 68.
1284 JH/10 The smtp_receive_timeout main option is now expanded before use.
1286 JH/11 The incoming_interface log option now also enables logging of the
1287 local interface on delivery outgoing connections.
1289 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
1290 if the interface and destination host and port all match.
1292 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
1295 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
1296 Patch from Andrew Lewis.
1298 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
1299 now supports optional time-restrictions, weighting, and priority
1300 modifiers per server. Patch originally by <rommer@active.by>.
1302 JH/16 The spamd_address main option now supports a mixed list of local
1303 and remote servers. Remote servers can be IPv6 addresses, and
1304 specify a port-range.
1306 JH/17 Bug 68: The spamd_address main option now supports an optional
1307 timeout value per server.
1309 JH/18 Bug 1581: Router and transport options headers_add/remove can
1310 now have the list separator specified.
1312 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
1315 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
1318 JH/21 Support for the A6 type of dns record is withdrawn.
1320 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
1321 rather than the verbs used.
1323 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
1324 from 255 to 1024 chars.
1326 JH/24 Verification callouts now attempt to use TLS by default.
1328 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
1329 are generic router options now. The defaults didn't change.
1331 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
1332 Original patch from Alexander Shikoff, worked over by JH.
1334 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
1335 files if ZCAT_COMMAND is not executable.
1337 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
1339 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
1341 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
1342 Normally benign, it bites when the pair was led to by a CNAME;
1343 modern usage is to not canonicalize the domain to a CNAME target
1344 (and we were inconsistent anyway for A-only vs AAAA+A).
1346 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
1348 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
1349 when evaluating $sender_host_dnssec.
1351 JH/31 Check the HELO verification lookup for DNSSEC, adding new
1352 $sender_helo_dnssec variable.
1354 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
1356 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
1358 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
1360 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
1361 documented as working, but never had. Support all but $spam_report.
1363 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
1364 added for tls authenticator.
1366 HS/03 Add perl_taintmode main config option
1372 TL/01 When running the test suite, the README says that variables such as
1373 no_msglog_check are global and can be placed anywhere in a specific
1374 test's script, however it was observed that placement needed to be near
1375 the beginning for it to behave that way. Changed the runtest perl
1376 script to read through the entire script once to detect and set these
1377 variables, reset to the beginning of the script, and then run through
1378 the script parsing/test process like normal.
1380 TL/02 The BSD's have an arc4random API. One of the functions to induce
1381 adding randomness was arc4random_stir(), but it has been removed in
1382 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
1383 function when detected.
1385 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
1386 cause callback expansion.
1388 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
1389 syntax errors in an expansion can be treated as a string instead of
1390 logging or causing an error, due to the internal use of bool_lax
1391 instead of bool when processing it.
1393 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
1394 server certificates when making smtp deliveries.
1396 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
1398 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
1400 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
1402 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
1403 Merged patch from Sebastian Wiedenroth.
1405 JH/05 Fix results-pipe from transport process. Several recipients, combined
1406 with certificate use, exposed issues where response data items split
1407 over buffer boundaries were not parsed properly. This eventually
1408 resulted in duplicates being sent. This issue only became common enough
1409 to notice due to the introduction of connection certificate information,
1410 the item size being so much larger. Found and fixed by Wolfgang Breyha.
1412 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
1413 size buffer was used, resulting in syntax errors when an expansion
1416 JH/07 Add support for directories of certificates when compiled with a GnuTLS
1417 version 3.3.6 or later.
1419 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
1420 is EXPERIMENTAL_EVENT, the main-configuration and transport options
1421 both become "event_action", the variables become $event_name, $event_data
1422 and $event_defer_errno. There is a new variable $verify_mode, usable in
1423 routers, transports and related events. The tls:cert event is now also
1424 raised for inbound connections, if the main configuration event_action
1427 TL/06 In test suite, disable OCSP for old versions of openssl which contained
1428 early OCSP support, but no stapling (appears to be less than 1.0.0).
1430 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
1431 server certificate names available under the smtp transport option
1432 "tls_verify_cert_hostname" now do not permit multi-component wildcard
1435 JH/10 Time-related extraction expansions from certificates now use the main
1436 option "timezone" setting for output formatting, and are consistent
1437 between OpenSSL and GnuTLS compilations. Bug 1541.
1439 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
1440 encoded parameter in the incoming message. Bug 1558.
1442 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
1443 include certificate info, eximon was claiming there were spoolfile
1446 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
1448 JH/14 Log delivery-related information more consistently, using the sequence
1449 "H=<name> [<ip>]" wherever possible.
1451 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
1452 are problematic for Debian distribution, omit them from the release
1455 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
1457 JH/16 Fix string representation of time values on 64bit time_t architectures.
1460 JH/17 Fix a null-indirection in certextract expansions when a nondefault
1461 output list separator was used.
1466 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
1467 checkers that were complaining about end of non-void function with no
1470 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
1471 This was a regression introduced in 4.83 by another bugfix.
1473 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1475 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1476 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1482 TF/01 Correctly close the server side of TLS when forking for delivery.
1484 When a message was received over SMTP with TLS, Exim failed to clear up
1485 the incoming connection properly after forking off the child process to
1486 deliver the message. In some situations the subsequent outgoing
1487 delivery connection happened to have the same fd number as the incoming
1488 connection previously had. Exim would try to use TLS and fail, logging
1489 a "Bad file descriptor" error.
1491 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1492 utilities have not been installed.
1494 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1495 temporary space as the ACL may create new global variables.
1497 TL/01 LDAP support uses per connection or global context settings, depending
1498 upon the detected version of the libraries at build time.
1500 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1501 to extract and use the src ip:port in logging and expansions as if it
1502 were a direct connection from the outside internet. PPv2 support was
1503 updated based on HAProxy spec change in May 2014.
1505 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1507 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1508 Properly escape header and check for NULL return.
1510 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1513 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1515 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1516 characters in header names, implemented as a verify condition.
1517 Contributed by Michael Fischer v. Mollard.
1519 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1520 results permerror and temperror. Previous values are deprecated but
1521 still accepted. In a future release, err_perm and err_temp will be
1522 completely removed, which will be a backward incompatibility if the
1523 ACL tests for either of these two old results. Patch contributed by
1524 user bes-internal on the mailing list.
1526 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1528 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1529 selectors, in both main and reject logs.
1531 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1534 JH/07 Add malware type "sock" for talking to simple daemon.
1536 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1538 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1539 routers/transports under cutthrough routing.
1541 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1542 numbers. Touch up "bool" conditional to keep the same definition.
1544 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1546 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1548 JH/12 Expand items in router/transport headers_add or headers_remove lists
1549 individually rather than the list as a whole. Bug 1452.
1551 Required for reasonable handling of multiple headers_ options when
1552 they may be empty; requires that headers_remove items with embedded
1553 colons must have them doubled (or the list-separator changed).
1555 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1556 view the policy declared in the DMARC record. Currently, $dmarc_status
1557 is a combined value of both the record presence and the result of the
1560 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1562 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1563 dnslookup router and the smtp transport (applying to the forward
1566 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1567 of ldap servers used for a specific lookup. Patch provided by Heiko
1570 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1571 New variable $lookup_dnssec_authenticated for observability.
1573 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1574 Patch submitted by Lars Timman.
1576 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1578 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1579 Requires trusted mode and valid format message id, aborts otherwise.
1580 Patch contributed by Heiko Schlichting.
1582 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1583 certextract with support for various fields. Bug 1358.
1585 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1586 is requested by default, modifiable by smtp transport option
1589 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1590 operate on certificate variables to give certificate fingerprints
1591 Also new ${sha256:cert_variable}.
1593 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1595 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1596 Christian Aistleitner.
1598 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1600 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1601 file. Patch from Wolfgang Breyha.
1603 JH/25 Expand the coverage of the delivery $host and $host_address to
1604 client authenticators run in verify callout. Bug 1476.
1606 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1607 align with daemon_smtp_ports. Bug 72.
1609 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1610 support and error reporting did not work properly.
1612 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1613 and is readable. Patch from Andrew Colin Kissa.
1615 TL/14 Enhance documentation of ${run expansion and how it parses the
1616 commandline after expansion, particularly in the case when an
1617 unquoted variable expansion results in an empty value.
1619 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1621 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1622 by renaming away from it. Addresses GH issue 15, reported by
1625 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1627 TL/15 SECURITY: prevent double expansion in math comparison functions
1628 (can expand unsanitized data). Not remotely exploitable.
1635 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1637 PP/02 Make -n do something, by making it not do something.
1638 When combined with -bP, the name of an option is not output.
1640 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1643 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1644 $sender_host_name and config options to manage this, and basic check
1647 PP/05 DSCP support for outbound connections and control modifier for inbound.
1649 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1650 (Only plugin which currently uses this is kerberos4, which nobody should
1651 be using, but we should make it available and other future plugins might
1652 conceivably use it, even though it would break NAT; stuff *should* be
1653 using channel bindings instead).
1655 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1656 name; added for Sendmail compatibility; requires admin caller.
1657 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1658 just ignore it); requires trusted caller.
1659 Also parse but ignore: -Ac -Am -X<logfile>
1662 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1664 TL/02 Add +smtp_confirmation as a default logging option.
1666 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1667 Patch by Magnus Holmgren from 2007-02-20.
1669 TL/04 Bugzilla 1281 - Spec typo.
1670 Bugzilla 1283 - Spec typo.
1671 Bugzilla 1290 - Spec grammar fixes.
1673 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1675 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1677 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1678 mailing list by by Dmitry Isaikin.
1680 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1682 JH/02 Support "G" suffix to numbers in ${if comparisons.
1684 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1686 NM/01 Bugzilla 1197 - Spec typo
1687 Bugzilla 1196 - Spec examples corrections
1689 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1691 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1692 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1695 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1696 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1698 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1699 "acl {{name}{arg}...}", and optional args on acl condition
1702 JH/05 Permit multiple router/transport headers_add/remove lines.
1704 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1706 JH/07 Avoid using a waiting database for a single-message-only transport.
1707 Performance patch from Paul Fisher. Bugzilla 1262.
1709 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1712 JH/09 Add $headers_added variable, with content from use of ACL modifier
1713 add_header (but not yet added to the message). Bugzilla 199.
1715 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1716 Pulled from Bugzilla 817 by Wolfgang Breyha.
1718 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1720 (nb: this is the same fix as in Exim 4.80.1)
1722 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1725 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1728 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1730 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1731 advertise SMTP AUTH mechanism to us, instead of a generic
1732 protocol violation error. Also, make Exim more robust to bad
1733 data from the Dovecot auth socket.
1735 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1737 When a queue runner is handling a message, Exim first routes the
1738 recipient addresses, during which it prunes them based on the retry
1739 hints database. After that it attempts to deliver the message to
1740 any remaining recipients. It then updates the hints database using
1743 So if a recipient address works intermittently, it can get repeatedly
1744 deferred at routing time. The retry hints record remains fresh so the
1745 address never reaches the final cutoff time.
1747 This is a fairly common occurrence when a user is bumping up against
1748 their storage quota. Exim had some logic in its local delivery code
1749 to deal with this. However it did not apply to per-recipient defers
1750 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1752 This change adds a proper retry rule check during routing so that the
1753 final cutoff time is checked against the message's age. We only do
1754 this check if there is an address retry record and there is not a
1755 domain retry record; this implies that previous attempts to handle
1756 the address had the retry_use_local_parts option turned on. We use
1757 this as an approximation for the destination being like a local
1758 delivery, as in LMTP.
1760 I suspect this new check makes the old local delivery cutoff check
1761 redundant, but I have not verified this so I left the code in place.
1763 TF/02 Correct gecos expansion when From: is a prefix of the username.
1765 Test 0254 submits a message to Exim with the header
1769 When I ran the test suite under the user fanf2, Exim expanded
1770 the header to contain my full name, whereas it should have added
1771 a Resent-Sender: header. It erroneously treats any prefix of the
1772 username as equal to the username.
1774 This change corrects that bug.
1776 GF/01 DCC debug and logging tidyup
1777 Error conditions log to paniclog rather than rejectlog.
1778 Debug lines prefixed by "DCC: " to remove any ambiguity.
1780 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1782 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1783 Bug spotted by Jeremy Harris; was flawed since initial commit.
1784 Would have resulted in OCSP responses post-SNI triggering an Exim
1785 NULL dereference and crash.
1787 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1789 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1790 Bug detection, analysis and fix by Samuel Thibault.
1791 Bugzilla 1331, Debian bug #698092.
1793 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1795 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1796 Server implementation by Todd Lyons, client by JH.
1797 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1798 config variable "prdr_enable" controls whether the server
1799 advertises the facility. If the client requests PRDR a new
1800 acl_data_smtp_prdr ACL is called once for each recipient, after
1801 the body content is received and before the acl_smtp_data ACL.
1802 The client is controlled by both of: a hosts_try_prdr option
1803 on the smtp transport, and the server advertisement.
1804 Default client logging of deliveries and rejections involving
1805 PRDR are flagged with the string "PRDR".
1807 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1808 fclose(). Diagnosis by Todd Lyons.
1810 PP/17 Update configure.default to handle IPv6 localhost better.
1811 Patch by Alain Williams (plus minor tweaks).
1814 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1815 This is now consistent with GnuTLS, and is now documented: the
1816 previous undocumented portable approach to treating the option as
1817 unset was to force an expansion failure. That still works, and
1818 an empty string is now equivalent.
1820 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1821 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1822 not performing validation itself.
1824 PP/20 Added force_command boolean option to pipe transport.
1825 Patch from Nick Koston, of cPanel Inc.
1827 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1830 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1832 PP/21 Fix eximon continuous updating with timestamped log-files.
1833 Broken in a format-string cleanup in 4.80, missed when I repaired the
1834 other false fix of the same issue.
1835 Report and fix from Heiko Schlichting.
1838 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1839 Report from Prashanth Katuri.
1841 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1842 It's SecureTransport, so affects any MacOS clients which use the
1843 system-integrated TLS libraries, including email clients.
1845 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1846 using a MIME ACL for non-SMTP local injection.
1847 Report and assistance in diagnosis by Warren Baker.
1849 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1851 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1853 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1854 last id that failed so it may be referenced in subsequent ACL's.
1856 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1859 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1860 ldap library initialization, allowing self-signed CA's to be
1861 used. Also properly sets require_cert option later in code by
1862 using NULL (global ldap config) instead of ldap handle (per
1863 session). Bug diagnosis and testing by alxgomz.
1865 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1866 the src/util/ subdirectory.
1868 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1869 renamed to Transport Post Delivery Action by Jeremy Harris, as
1872 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1873 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1874 redis_servers = needs to be configured which will be used by the redis
1875 lookup. Patch from Warren Baker, of The Packet Hub.
1877 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1879 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1880 hostname or reverse DNS when processing a host list. Used suggestions
1881 from multiple comments on this bug.
1883 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1885 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1886 Missed a few lines, added it to make the runtest require no keyboard
1889 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1890 contains upper case chars. Make router use caseful_local_part.
1892 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1893 support when GnuTLS has been built with p11-kit.
1899 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1901 This, or similar/improved, will also be change PP/11 of 4.82.
1907 PP/01 Handle short writes when writing local log-files.
1908 In practice, only affects FreeBSD (8 onwards).
1909 Bugzilla 1053, with thanks to Dmitry Isaikin.
1911 NM/01 Bugzilla 949 - Documentation tweak
1913 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1916 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1918 PP/02 Implemented gsasl authenticator.
1920 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1922 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1923 `pkg-config foo` for cflags/libs.
1925 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1926 with rest of GSASL and with heimdal_gssapi.
1928 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1929 `pkg-config foo` for cflags/libs for the TLS implementation.
1931 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1932 properties get this fed in as external SSF. A number of robustness
1933 and debugging improvements to the cyrus_sasl authenticator.
1935 PP/08 cyrus_sasl server now expands the server_realm option.
1937 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1938 Patch by Jeremy Harris.
1940 PP/10 Added dbmjz lookup type.
1942 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1944 PP/12 MAIL args handles TAB as well as SP, for better interop with
1945 non-compliant senders.
1946 Analysis and variant patch by Todd Lyons.
1948 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1949 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1950 Patch from Dirk Mueller <dmueller@suse.com>
1952 PP/13 tls_peerdn now print-escaped for spool files.
1953 Observed some $tls_peerdn in wild which contained \n, which resulted
1954 in spool file corruption.
1956 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1957 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1958 or write after TLS renegotiation, which otherwise led to messages
1961 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1962 as a tracking header (ie: a signed header comes before the signature).
1963 Patch from Wolfgang Breyha.
1965 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1966 comma-sep list; embedded commas doubled.
1968 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1970 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1972 Report and patch from Dmitry Banschikov.
1974 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1975 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1976 failures. We appear to now support TLS1.1+ with Exim.
1978 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1979 lets Exim select keys and certificates based upon TLS SNI from client.
1980 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1981 before an outbound SMTP session. New log_selector, +tls_sni.
1983 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1984 NULL dereference. Report and patch from Alun Jones.
1986 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1987 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1988 Not seeing resolver debug output on NetBSD, but suspect this is a
1989 resolver implementation change.
1991 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1992 Left warnings. Added "eximon gdb" invocation mode.
1994 PP/21 Defaulting "accept_8bitmime" to true, not false.
1996 PP/22 Added -bw for inetd wait mode support.
1998 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1999 locate the relevant includes and libraries. Made this the default.
2001 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
2002 Bugzilla 1246, report and most of solution from Tomasz Kusy.
2004 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
2005 This may cause build issues on older platforms.
2007 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
2008 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
2009 gnutls_require_mac & gnutls_require_protocols (no longer supported).
2010 Added SNI support via GnuTLS too.
2011 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
2013 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
2015 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
2016 Applied second patch from Janne, implementing suggestion to default
2017 multiple-strings-in-record handling to match SPF spec.
2019 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
2021 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
2022 read-only, out of scope).
2023 Patch from Wolfgang Breyha, report from Stuart Northfield.
2025 PP/29 Fix three issues highlighted by clang analyser static analysis.
2026 Only crash-plausible issue would require the Cambridge-specific
2027 iplookup router and a misconfiguration.
2028 Report from Marcin Mirosław.
2030 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
2032 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
2033 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
2034 As part of this, removing so much warning spew let me fix some minor
2035 real issues in debug logging.
2037 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
2038 assignment on my part. Fixed.
2040 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
2041 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
2042 Janne Snabb (who went above and beyond: thank you).
2044 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
2045 string otherwise requires a connection and a bunch more work and it's
2046 relatively easy to get wrong. Should also expose TLS library linkage
2049 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
2050 64-bit ${eval} (JH/03).
2052 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
2053 GNU libc to support some of the 64-bit stuff, should not lead to
2054 conflicts. Defined before os.h is pulled in, so if a given platform
2055 needs to override this, it can.
2057 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
2058 protection layer was required, which is not implemented.
2059 Bugzilla 1254, patch from Wolfgang Breyha.
2061 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
2062 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
2063 tls_dhparam take prime identifiers. Also unbreak combination of
2064 OpenSSL+DH_params+TLSSNI.
2066 PP/39 Disable SSLv2 by default in OpenSSL support.
2072 PP/01 Solaris build fix for Oracle's LDAP libraries.
2073 Bugzilla 1109, patch from Stephen Usher.
2075 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
2077 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
2080 TF/02 Fix a couple more cases where we did not log the error message
2081 when unlink() failed. See also change 4.74-TF/03.
2083 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
2084 lock up or crash if it happened to be inside a call to libc when it
2085 got a SIGUSR1 from exiwhat.
2087 The SIGUSR1 handler appends the current process status to the process
2088 log which is later printed by exiwhat. It used to use the general
2089 purpose logging code to do this, but several functions it calls are
2090 not safe for signals.
2092 The new output code in the SIGUSR1 handler is specific to the process
2093 log, and simple enough that it's easy to inspect for signal safety.
2094 Removing some special cases also simplifies the general logging code.
2095 Removing the spurious timestamps from the process log simplifies
2098 TF/04 Improved ratelimit ACL condition.
2100 The /noupdate option has been deprecated in favour of /readonly which
2101 has clearer semantics. The /leaky, /strict, and /readonly update modes
2102 are mutually exclusive. The update mode is no longer included in the
2103 database key; it just determines when the database is updated. (This
2104 means that when you upgrade Exim will forget old rate measurements.)
2106 Exim now checks that the per_* options are used with an update mode that
2107 makes sense for the current ACL. For example, when Exim is processing a
2108 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
2109 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
2110 must specify per_mail/readonly. If you omit the update mode it defaults to
2111 /leaky where that makes sense (as before) or /readonly where required.
2113 The /noupdate option is now undocumented but still supported for
2114 backwards compatibility. It is equivalent to /readonly except that in
2115 ACLs where /readonly is required you may specify /leaky/noupdate or
2116 /strict/noupdate which are treated the same as /readonly.
2118 A useful new feature is the /count= option. This is a generalization
2119 of the per_byte option, so that you can measure the throughput of other
2120 aggregate values. For example, the per_byte option is now equivalent
2121 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
2123 The per_rcpt option has been generalized using the /count= mechanism
2124 (though it's more complicated than the per_byte equivalence). When it is
2125 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
2126 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
2127 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
2128 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
2129 Note that using per_rcpt with a non-readonly update mode in more than
2130 one ACL will cause the recipients to be double-counted. (The per_mail
2131 and per_byte options don't have this problem.)
2133 The handling of very low rates has changed slightly. If the computed rate
2134 is less than the event's count (usually one) then this event is the first
2135 after a long gap. In this case the rate is set to the same as this event's
2136 count, so that the first message of a spam run is counted properly.
2138 The major new feature is a mechanism for counting the rate of unique
2139 events. The new per_addr option counts the number of different
2140 recipients that someone has sent messages to in the last time period. It
2141 behaves like per_rcpt if all the recipient addresses are different, but
2142 duplicate recipient addresses do not increase the measured rate. Like
2143 the /count= option this is a general mechanism, so the per_addr option
2144 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
2145 example, measure the rate that a client uses different sender addresses
2146 with the options per_mail/unique=$sender_address. There are further
2147 details in the main documentation.
2149 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
2151 TF/06 Removed a few PCRE remnants.
2153 TF/07 Automatically extract Exim's version number from tags in the git
2154 repository when doing development or release builds.
2156 PP/02 Raise smtp_cmd_buffer_size to 16kB.
2157 Bugzilla 879. Patch from Paul Fisher.
2159 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
2160 Heavily based on revision 40f9a89a from Simon Arlott's tree.
2163 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
2165 PP/05 Variable $av_failed, true if the AV scanner deferred.
2166 Bugzilla 1078. Patch from John Horne.
2168 PP/06 Stop make process more reliably on build failure.
2169 Bugzilla 1087. Patch from Heiko Schlittermann.
2171 PP/07 Make maildir_use_size_file an _expandable_ boolean.
2172 Bugzilla 1089. Patch from Heiko Schlittermann.
2174 PP/08 Handle ${run} returning more data than OS pipe buffer size.
2175 Bugzilla 1131. Patch from Holger Weiß.
2177 PP/09 Handle IPv6 addresses with SPF.
2178 Bugzilla 860. Patch from Wolfgang Breyha.
2180 PP/10 GnuTLS: support TLS 1.2 & 1.1.
2182 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
2185 PP/11 match_* no longer expand right-hand-side by default.
2186 New compile-time build option, EXPAND_LISTMATCH_RHS.
2187 New expansion conditions, "inlist", "inlisti".
2189 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
2191 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
2193 PP/14 fix log_write() format string regression from TF/03.
2194 Bugzilla 1152. Patch from Dmitry Isaikin.
2200 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
2202 PP/02 Harmonised TLS library version reporting; only show if debugging.
2203 Layout now matches that introduced for other libraries in 4.74 PP/03.
2205 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
2207 PP/04 New "dns_use_edns0" global option.
2209 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
2212 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
2213 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
2215 TK/01 Updated PolarSSL code to 0.14.2.
2216 Bugzilla 1097. Patch from Andreas Metzler.
2218 PP/07 Catch divide-by-zero in ${eval:...}.
2219 Fixes bugzilla 1102.
2221 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
2224 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
2225 format-string attack -- SECURITY: remote arbitrary code execution.
2227 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
2228 time unintentionally subject to list matching rules, letting the header
2229 cause arbitrary Exim lookups (of items which can occur in lists, *not*
2230 arbitrary string expansion). This allowed for information disclosure.
2232 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
2233 INT_MIN/-1 -- value coerced to INT_MAX.
2239 NM/01 Workaround for PCRE version dependency in version reporting
2242 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
2243 This fixes portability to compilers other than gcc, notably
2244 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
2246 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
2247 makefiles for portability to HP-UX and POSIX correctness.
2249 PP/01 Permit LOOKUP_foo enabling on the make command-line.
2250 Also via indented variable definition in the Makefile.
2251 (Debugging by Oliver Heesakkers).
2253 PP/02 Restore caching of spamd results with expanded spamd_address.
2254 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
2256 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
2257 Improves build reliability. Fix from: Frank Elsner
2259 NM/02 Fix wide character breakage in the rfc2047 coding
2260 Fixes bug 1064. Patch from Andrey N. Oktyabrski
2262 NM/03 Allow underscore in dnslist lookups
2263 Fixes bug 1026. Patch from Graeme Fowler
2265 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
2266 Code patches from Adam Ciarcinski of NetBSD.
2268 NM/04 Fixed exiqgrep to cope with mailq missing size issue
2271 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
2272 is logged, to avoid truncation. Patch from John Horne.
2274 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
2275 Patch from Jakob Hirsch.
2277 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
2278 SQL string expansion failure details.
2279 Patch from Andrey Oktyabrski.
2281 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
2282 Patch from Simon Arlott.
2284 PP/09 New lookups functionality failed to compile on old gcc which rejects
2285 extern declarations in function scope.
2286 Patch from Oliver Fleischmann
2288 PP/10 Use sig_atomic_t for flags set from signal handlers.
2289 Check getgroups() return and improve debugging.
2290 Fixed developed for diagnosis in bug 927 (which turned out to be
2293 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
2294 Patch from Mark Zealey.
2296 PP/12 Bugzilla 1056: Improved spamd server selection.
2297 Patch from Mark Zealey.
2299 PP/13 Bugzilla 1086: Deal with maildir quota file races.
2300 Based on patch from Heiko Schlittermann.
2302 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
2303 Patch from Uwe Doering, sign-off by Michael Haardt.
2305 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
2306 variable declaration deep within a block. Bug and patch from
2309 PP/15 lookups-Makefile IRIX compatibility coercion.
2311 PP/16 Make DISABLE_DKIM build knob functional.
2313 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
2314 Patch by Simon Arlott
2316 TF/03 Fix valgrind.h portability to C89 compilers that do not support
2317 variable argument macros. Our copy now differs from upstream.
2323 TF/01 Failure to get a lock on a hints database can have serious
2324 consequences so log it to the panic log.
2326 TF/02 Log LMTP confirmation messages in the same way as SMTP,
2327 controlled using the smtp_confirmation log selector.
2329 TF/03 Include the error message when we fail to unlink a spool file.
2331 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
2332 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
2333 for maintaining out-of-tree patches for some time.
2335 PP/01 Bugzilla 139: Documentation and portability issues.
2336 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
2337 Handle per-OS dynamic-module compilation flags.
2339 PP/02 Let /dev/null have normal permissions.
2340 The 4.73 fixes were a little too stringent and complained about the
2341 permissions on /dev/null. Exempt it from some checks.
2342 Reported by Andreas M. Kirchwitz.
2344 PP/03 Report version information for many libraries, including
2345 Exim version information for dynamically loaded libraries. Created
2346 version.h, now support a version extension string for distributors
2347 who patch heavily. Dynamic module ABI change.
2349 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
2350 privilege escalation vulnerability whereby the Exim run-time user
2351 can cause root to append content of the attacker's choosing to
2354 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
2357 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
2358 If dropping privileges for untrusted macros, we disabled normal logging
2359 on the basis that it would fail; for the Exim run-time user, this is not
2360 the case, and it resulted in successful deliveries going unlogged.
2361 Fixed. Reported by Andreas Metzler.
2367 PP/01 Date: & Message-Id: revert to normally being appended to a message,
2368 only prepend for the Resent-* case. Fixes regression introduced in
2369 Exim 4.70 by NM/22 for Bugzilla 607.
2371 PP/02 Include check_rfc2047_length in configure.default because we're seeing
2372 increasing numbers of administrators be bitten by this.
2374 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
2376 PP/03 Bugzilla 994: added openssl_options main configuration option.
2378 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
2380 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
2382 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
2384 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
2385 without a peer certificate, leading to a segfault because of an
2386 assumption that peers always have certificates. Be a little more
2387 paranoid. Problem reported by Martin Tscholak.
2389 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
2390 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
2391 NB: ClamAV planning to remove STREAM in "middle of 2010".
2392 CL also introduces -bmalware, various -d+acl logging additions and
2393 more caution in buffer sizes.
2395 PP/09 Implemented reverse_ip expansion operator.
2397 PP/10 Bugzilla 937: provide a "debug" ACL control.
2399 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
2401 PP/12 Bugzilla 973: Implement --version.
2403 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
2405 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
2407 PP/15 Bugzilla 816: support multiple condition rules on Routers.
2409 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
2410 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
2411 ignore trailing whitespace.
2413 JJ/02 prevent non-panic DKIM error from being sent to paniclog
2415 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
2418 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
2419 Notification from Dr Andrew Aitchison.
2421 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
2422 ExtendedDetectionInfo response format.
2423 Notification from John Horne.
2425 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
2428 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
2429 XSL and documented dependency on system catalogs, with examples of how
2432 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
2435 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
2436 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
2437 configuration file which is writeable by the Exim user or group.
2439 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
2440 of configuration files to cover files specified with the -C option if
2441 they are going to be used with root privileges, not just the default
2444 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
2445 option (effectively making it always true).
2447 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
2448 files to be used while preserving root privileges.
2450 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
2451 that rogue child processes cannot use them.
2453 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
2454 run-time user, instead of root.
2456 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
2457 Exim run-time user without dropping privileges.
2459 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
2460 result string, instead of calling string_vformat() twice with the same
2463 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
2464 for other users. Others should always drop root privileges if they use
2465 -C on the command line, even for a whitelisted configure file.
2467 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
2469 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2475 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2476 $data_path, and $header_path variables; fixed documentation bugs and
2479 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2480 exipick to access non-standard spools, including the "frozen" queue
2483 NM/01 Bugzilla 965: Support mysql stored procedures.
2484 Patch from Alain Williams
2486 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2488 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2489 Patch from Andreas Metzler
2491 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2492 Patch from Kirill Miazine
2494 NM/05 Bugzilla 671: Added umask to procmail example.
2496 JJ/03 installed exipick 20100323.0, fixing doc bug
2498 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2499 directory. Notification and patch from Dan Rosenberg.
2501 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2503 TK/02 Improve log output when DKIM signing operation fails.
2505 MH/01 Treat the transport option dkim_domain as a colon separated
2506 list, not as a single string, and sign the message with each element,
2507 omitting multiple occurences of the same signer.
2509 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2510 Bugzilla 985, 986. Patch by Simon Arlott
2512 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2513 Patch by Simon Arlott
2515 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2516 MBX locking. Notification from Dan Rosenberg.
2522 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2524 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2526 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2528 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2530 NM/04 Bugzilla 915: Flag broken perl installation during build.
2536 TK/01 Added patch by Johannes Berg that expands the main option
2537 "spamd_address" if it starts with a dollar sign.
2539 TK/02 Write list of recipients to X-Envelope-Sender header when building
2540 the mbox-format spool file for content scanning (suggested by Jakob
2543 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2544 (http://www.dcc-servers.net/) support via dccifd. Activated by
2545 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2547 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2548 by Mark Daniel Reidel <mr@df.eu>.
2550 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2551 When building exim an external PCRE library is now needed -
2552 PCRE is a system library on the majority of modern systems.
2553 See entry on PCRE_LIBS in EDITME file.
2555 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2556 conversation. Added nologin parameter to request.
2557 Patch contributed by Kirill Miazine.
2559 TF/01 Do not log submission mode rewrites if they do not change the address.
2561 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2563 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2564 log files in place. Contributed by Roberto Lima.
2566 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2568 TF/03 Bugzilla 615: When checking the local_parts router precondition
2569 after a local_part_suffix or local_part_prefix option, Exim now
2570 does not use the address's named list lookup cache, since this
2571 contains cached lookups for the whole local part.
2573 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2574 Robert Millan. Documentation is in experimental-spec.txt.
2576 TF/04 Bugzilla 668: Fix parallel build (make -j).
2578 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2580 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2581 Patch provided by Jan Srzednicki.
2583 TF/05 Leading white space used to be stripped from $spam_report which
2584 wrecked the formatting. Now it is preserved.
2586 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2587 that they are available at delivery time.
2589 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2591 TF/08 TLS error reporting now respects the incoming_interface and
2592 incoming_port log selectors.
2594 TF/09 Produce a more useful error message if an SMTP transport's hosts
2595 setting expands to an empty string.
2597 NM/06 Bugzilla 744: EXPN did not work under TLS.
2598 Patch provided by Phil Pennock.
2600 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2601 Patch provided by Richard Godbee.
2603 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2604 acl_smtp_notquit, added index entry.
2606 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2607 Patch provided by Eugene Bujak.
2609 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2610 accept(). Patch provided by Maxim Dounin.
2612 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2613 Patch provided by Phil Pennock.
2615 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2617 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2618 Patch provided by Brad "anomie" Jorsch.
2620 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2621 Patch provided by Dean Brooks.
2623 TK/05 Add native DKIM support (does not depend on external libraries).
2625 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2626 Patch provided by Graeme Fowler.
2628 NM/16 Bugzilla 851: Documentation example syntax fix.
2630 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2632 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2635 NM/19 Bugzilla 745: TLS version reporting.
2636 Patch provided by Phil Pennock.
2638 NM/20 Bugzilla 167: bool: condition support.
2639 Patch provided by Phil Pennock.
2641 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2642 clients. Patch provided by Phil Pennock.
2644 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2645 Patch provided by Brad "anomie" Jorsch.
2647 NM/23 Bugzilla 687: Fix misparses in eximstats.
2648 Patch provided by Heiko Schlittermann.
2650 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2651 Patch provided by Heiko Schlittermann.
2653 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2654 plus update to original patch.
2656 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2658 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2659 Patch provided by David Brownlee.
2661 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2663 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2665 NM/30 Bugzilla 888: TLS documentation bugfixes.
2667 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2669 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2670 Unlike the original bugzilla I have changed all shell scripts in src tree.
2672 NM/33 Bugzilla 898: Transport filter timeout fix.
2673 Patch by Todd Rinaldo.
2675 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2676 Patch by Serge Demonchaux.
2678 NM/35 Bugzilla 39: Base64 decode bug fixes.
2679 Patch by Jakob Hirsch.
2681 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2683 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2685 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2687 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2693 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2694 ALT-N's libdkim that I have put here:
2695 http://duncanthrax.net/exim-experimental/
2697 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2698 were called 'true' and it seems that C99 defines that as a reserved
2699 keyword to be used with 'bool' variable types. That means you could
2700 not include C99-style headers which use bools without triggering
2701 build errors in sieve.c.
2703 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2704 as mailq or other aliases. Changed the --help handling significantly
2705 to do whats expected. exim_usage() emits usage/help information.
2707 SC/01 Added the -bylocaldomain option to eximstats.
2709 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2711 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2713 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2719 PH/01 Another patch from the Sieve maintainer.
2721 PH/02 When an IPv6 address is converted to a string for single-key lookup
2722 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2723 dots are used instead of colons so that keys in lsearch files need not
2724 contain colons. This was done some time before quoting was made available
2725 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2726 (notated using the quote facility) so as to distinguish them from IPv4
2727 keys. This meant that lookups for IP addresses in host lists did not work
2728 for iplsearch lookups.
2730 This has been fixed by arranging for IPv6 addresses to be expressed with
2731 colons if the lookup type is iplsearch. This is not incompatible, because
2732 previously such lookups could never work.
2734 The situation is now rather anomalous, since one *can* have colons in
2735 ordinary lsearch keys. However, making the change in all cases is
2736 incompatible and would probably break a number of configurations.
2738 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2741 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2742 conversion specification without a maximum field width, thereby enabling
2743 a rogue spamd server to cause a buffer overflow. While nobody in their
2744 right mind would setup Exim to query an untrusted spamd server, an
2745 attacker that gains access to a server running spamd could potentially
2746 exploit this vulnerability to run arbitrary code as the Exim user.
2748 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2749 $primary_hostname instead of what libspf2 thinks the hosts name is.
2751 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2752 a directory entry by the name of the lookup key. Previously, if a
2753 symlink pointed to a non-existing file or a file in a directory that
2754 Exim lacked permissions to read, a lookup for a key matching that
2755 symlink would fail. Now it is enough that a matching directory entry
2756 exists, symlink or not. (Bugzilla 503.)
2758 PH/03 The body_linecount and body_zerocount variables are now exported in the
2761 PH/04 Added the $dnslist_matched variable.
2763 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2764 This means they are set thereafter only if the connection becomes
2767 PH/06 Added the client_condition to authenticators so that some can be skipped
2768 by clients under certain conditions.
2770 PH/07 The error message for a badly-placed control=no_multiline_responses left
2771 "_responses" off the end of the name.
2773 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2775 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2776 (without spaces) instead of just copying the configuration text.
2778 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2780 PH/11 Added $max_received_linelength.
2782 PH/12 Added +ignore_defer and +include_defer to host lists.
2784 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2785 way in which PCRE > 7.0 is built.
2787 PH/14 Implemented queue_only_load_latch.
2789 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2790 MAIL command. The effect was to mangle the value on 64-bit systems.
2792 PH/16 Another patch from the Sieve maintainer.
2794 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2796 PH/18 If a system quota error occurred while trying to create the file for
2797 a maildir delivery, the message "Mailbox is full" was not appended to the
2798 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2799 only to a quota excession during the actual writing of the file.
2801 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2802 characters?) which causes problems in log lines. The DN values are now
2803 passed through string_printing() before being added to log lines.
2805 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2806 and InterBase are left for another time.)
2808 PH/21 Added message_body_newlines option.
2810 PH/22 Guard against possible overflow in moan_check_errorcopy().
2812 PH/23 POSIX allows open() to be a macro; guard against that.
2814 PH/24 If the recipient of an error message contained an @ in the local part
2815 (suitably quoted, of course), incorrect values were put in $domain and
2816 $local_part during the evaluation of errors_copy.
2822 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2823 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2826 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2827 issue a MAIL command.
2829 PH/02 In an ACL statement such as
2831 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2833 if a client was not listed at all, or was listed with a value other than
2834 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2835 the condition was not true (as it should be), so access was not denied.
2836 The bug was that the ! inversion was incorrectly passed on to the second
2837 item. This has been fixed.
2839 PH/03 Added additional dnslists conditions == and =& which are different from
2840 = and & when the dns lookup returns more than one IP address.
2842 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2843 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2845 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2846 FSYNC, which compiles an option called disable_fsync that allows for
2847 bypassing fsync(). The documentation is heavily laced with warnings.
2849 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2851 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2852 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2853 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2854 including adding "make clean"; (3) Added -fPIC when compiling the test
2855 dynamically loaded module, to get rid of a warning.
2857 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2858 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2859 = 0s. The bug is otherwise harmless.
2861 PH/07 There was a bug in the dovecot authenticator such that the value of
2862 $auth1 could be overwritten, and so not correctly preserved, after a
2863 successful authentication. This usually meant that the value preserved by
2864 the server_setid option was incorrect.
2866 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2868 PH/09 Installed PCRE release 7.0.
2870 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2871 run for batched SMTP input. It is now run at the start of every message
2872 in the batch. While fixing this I discovered that the process information
2873 (output by running exiwhat) was not always getting set for -bs and -bS
2874 input. This is fixed, and it now also says "batched" for BSMTP.
2876 PH/11 Added control=no_pipelining.
2878 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2879 patch, slightly modified), and move the expansion of helo_data till after
2880 the connection is made in the smtp transport (so it can use these
2883 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2885 PH/14 Added log_selector = +pid.
2887 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2889 PH/16 Add ${if forany and ${if forall.
2891 PH/17 Added dsn_from option to vary the From: line in DSNs.
2893 PH/18 Flush SMTP output before performing a callout, unless control =
2894 no_callout_flush is set.
2896 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2897 was true (the default) a successful delivery failed to delete the retry
2898 item, thus causing premature timeout of the address. The bug is now
2901 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2903 PH/21 Long custom messages for fakedefer and fakereject are now split up
2904 into multiline responses in the same way that messages for "deny" and
2905 other ACL rejections are.
2907 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2908 with slight modification.
2910 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2911 draft, changing the syntax and factoring some duplicate code".
2913 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2914 for deliveries of the second and subsequent messages over the same SMTP
2917 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2918 ${reduce, with only minor "tidies".
2920 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2922 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2923 expansion side effects.
2925 PH/27 When a message times out after an over-quota error from an Exim-imposed
2926 quota, the bounce message says "mailbox is full". This message was not
2927 being given when it was a system quota that was exceeded. It now should
2930 MH/03 Made $recipients available in local_scan(). local_scan() already has
2931 better access to the recipient list through recipients_list[], but
2932 $recipients can be useful in postmaster-provided expansion strings.
2934 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2935 in the case of a MAIL command with additional options following the
2936 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2937 were accidentally chopped off.
2939 PH/29 SMTP synchronization checks are implemented when a command is read -
2940 there is a check that no more input is waiting when there shouldn't be
2941 any. However, for some commands, a delay in an ACL can mean that it is
2942 some time before the response is written. In this time, more input might
2943 arrive, invalidly. So now there are extra checks after an ACL has run for
2944 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2945 pipelining has not been advertised.
2947 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2949 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2950 returned in the SMTP response when smtp_return_error_details was false.
2951 This has been fixed.
2953 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2954 instead of the C I/O that was originally supplied, because problems were
2955 reported on Solaris.
2957 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2958 Exim which did not show up earlier: it was assuming that a call to
2959 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2960 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2961 was a macro that became an assignment, so it seemed to work. This has
2962 changed to a proper function call with a void return, hence the compile
2963 error. Exim's code has been fixed.
2965 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2968 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2970 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2972 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2973 because it turns out to be too big or there is a timeout) there is no
2974 "Completed" line in the log. When some messages of this type were
2975 selected by exigrep, they were listed as "not completed". Others were
2976 picked up by some special patterns. I have improved the selection
2977 criteria to be more general.
2979 PH/38 The host_find_failed option in the manualroute router can now be set
2980 to "ignore", to completely ignore a host whose IP address cannot be
2981 found. If all hosts are ignored, the behaviour is controlled by the new
2982 host_all_ignored option.
2984 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2985 homing or because of multiple MX records with /mx) generated more than
2986 one IP address, and the following item turned out to be the local host,
2987 all the secondary addresses of the first item were incorrectly removed
2988 from the list, along with the local host and any following hosts (which
2989 is what is supposed to happen).
2991 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2992 whoever called Exim into the -H file. In the case of the daemon it was
2993 behaving confusingly. When first started, it used values for whoever
2994 started the daemon, but after a SIGHUP it used the Exim user (because it
2995 calls itself on a restart). I have changed the code so that it now always
2998 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2999 message are rejected with the same error (e.g. no authentication or bad
3000 sender address), and a DATA command is nevertheless sent (as can happen
3001 with PIPELINING or a stupid MUA), the error message that was given to the
3002 RCPT commands is included in the rejection of the DATA command. This is
3003 intended to be helpful for MUAs that show only the final error to their
3006 PH/42 Another patch from the Sieve maintainer.
3008 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
3009 Eximstats - Fixed some broken HTML links and added missing column headers
3011 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
3012 columns for Rejects, Temp Rejects, Ham, and Spam rows.
3014 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
3016 PH/43 Yet another patch from the Sieve maintainer.
3018 PH/44 I found a way to check for a TCP/IP connection going away before sending
3019 the response to the final '.' that terminates a message, but only in the
3020 case where the client has not sent further data following the '.'
3021 (unfortunately, this is allowed). However, in many cases there won't be
3022 any further data because there won't be any more messages to send. A call
3023 to select() can be used: if it shows that the input is "ready", there is
3024 either input waiting, or the socket has been closed. An attempt to read
3025 the next input character can distinguish the two cases. Previously, Exim
3026 would have sent an OK response which the client would never have see.
3027 This could lead to message repetition. This fix should cure that, at
3028 least in a lot of common cases.
3030 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
3031 advertised in response to EHLO.
3037 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
3038 fixed by 4.65/MH/01 (is this a record?) are fixed:
3040 (i) An empty string was always treated as zero by the numeric comparison
3041 operators. This behaviour has been restored.
3043 (ii) It is documented that the numeric comparison operators always treat
3044 their arguments as decimal numbers. This was broken in that numbers
3045 starting with 0 were being interpreted as octal.
3047 While fixing these problems I realized that there was another issue that
3048 hadn't been noticed. Values of message_size_limit (both the global option
3049 and the transport option) were treated as octal if they started with 0.
3050 The documentation was vague. These values are now always treated as
3051 decimal, and I will make that clear in the documentation.
3057 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
3058 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
3061 MH/01 Don't check that the operands of numeric comparison operators are
3062 integers when their expansion is in "skipping" mode (fixes bug
3063 introduced by 4.64-PH/07).
3065 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
3066 child addresses, Exim now panics and dies. Previously, because the count
3067 is held in a short int, deliveries were likely to be lost. As such a
3068 large number of recipients for a single message is ridiculous
3069 (performance will be very, very poor), I have chosen to impose a limit
3070 rather than extend the field.
3076 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
3077 leftover -K file (the existence of which was triggered by #402).
3078 While we were at it, introduced process PID as part of the -K
3079 filename. This should rule out race conditions when creating
3082 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
3083 processing considerably. Previous code took too long for large mails,
3084 triggering a timeout which in turn triggers #401.
3086 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
3087 in the DK code in transports.c. sendfile() is not really portable,
3088 hence the _LINUX specificness.
3090 TF/01 In the add_headers option to the mail command in an Exim filter,
3091 there was a bug that Exim would claim a syntax error in any
3092 header after the first one which had an odd number of characters
3095 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
3096 callout verification, Exim cached a "reject" for the entire domain. This
3097 is correct for most verifications, but it is not correct for a recipient
3098 verification with use_sender or use_postmaster set, because in that case
3099 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
3100 case of MAIL FROM:<> rejection from other early rejections (e.g.
3101 rejection of HELO). When verifying a recipient using a non-null MAIL
3102 address, the cache is ignored if it shows MAIL FROM:<> rejection.
3103 Whatever the result of the callout, the value of the domain cache is
3104 left unchanged (for any other kind of callout, getting as far as trying
3105 RCPT means that the domain itself is ok).
3107 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
3110 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
3111 manifest itself as EPIPE rather than ECONNECT. When tidying away a
3112 session, the daemon ignores ECONNECT errors and logs others; it now
3113 ignores EPIPE as well.
3115 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
3116 (quoted-printable decoding).
3118 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
3119 later the small subsequent patch to fix an introduced bug.
3121 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
3123 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
3125 PH/08 An error is now given if message_size_limit is specified negative.
3127 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
3128 to be given (somewhat) arbitrary names.
3130 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
3133 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
3134 miscellaneous code fixes
3136 PH/10 Added the log_reject_target ACL modifier to specify where to log
3139 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
3140 hostname. This is wrong, because it relates to the incoming message (and
3141 probably the interface on which it is arriving) and not to the outgoing
3142 callout (which could be using a different interface). This has been
3143 changed to use the value of the helo_data option from the smtp transport
3144 instead - this is what is used when a message is actually being sent. If
3145 there is no remote transport (possible with a router that sets up host
3146 addresses), $smtp_active_hostname is used.
3148 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
3149 tweaks were necessary in order to get it to work (see also 21 below):
3150 (a) The code assumed that strncpy() returns a negative number on buffer
3151 overflow, which isn't the case. Replaced with Exim's string_format()
3153 (b) There were several signed/unsigned issues. I just did the minimum
3154 hacking in of casts. There is scope for a larger refactoring.
3155 (c) The code used strcasecmp() which is not a standard C function.
3156 Replaced with Exim's strcmpic() function.
3157 (d) The code set only $1; it now sets $auth1 as well.
3158 (e) A simple test gave the error "authentication client didn't specify
3159 service in request". It would seem that Dovecot has changed its
3160 interface. Fortunately there's a specification; I followed it and
3161 changed what the client sends and it appears to be working now.
3163 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
3166 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
3167 address A is aliased to B and C, where B exists and C does not. Without
3168 -v the output is "A verified" because verification stops after a
3169 successful redirection if more than one address is generated. However,
3170 with -v the child addresses are also verified. Exim was outputting "A
3171 failed to verify" and then showing the successful verification for C,
3172 with its parentage. It now outputs "B failed to verify", showing B's
3173 parentage before showing the successful verification of C.
3175 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
3176 look up a TXT record in a specific list after matching in a combined
3179 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
3180 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
3181 they consult the DNS. I had assumed they would set it the way they
3182 wanted; and indeed my experiments on Linux seem to show that in some
3183 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
3184 To be on the safe side, however, I have now made the interface to
3185 host_find_byname() similar to host_find_bydns(), with an argument
3186 containing the DNS resolver options. The host_find_byname() function now
3187 sets these options at its start, just as host_find_bydns() does. The smtp
3188 transport options dns_qualify_single and dns_search_parents are passed to
3189 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
3190 of host_find_byname() use the default settings of RES_DEFNAMES
3191 (qualify_single) but not RES_DNSRCH (search_parents).
3193 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
3194 spool_read_header() do less string testing, by means of a preliminary
3195 switch on the second character of optional "-foo" lines. (This is
3196 overdue, caused by the large number of possibilities that now exist.
3197 Originally there were few.) While I was there, I also converted the
3198 str(n)cmp tests so they don't re-test the leading "-" and the first
3199 character, in the hope this might squeeze out yet more improvement.
3201 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
3202 flag allowing group syntax was set by the header_syntax check but not
3203 turned off, possible causing trouble later; (2) The flag was not being
3204 set at all for the header_verify test, causing "group"-style headers to
3205 be rejected. I have now set it in this case, and also caused header_
3206 verify to ignore an empty address taken from a group. While doing this, I
3207 came across some other cases where the code for allowing group syntax
3208 while scanning a header line wasn't quite right (mostly, not resetting
3209 the flag correctly in the right place). These bugs could have caused
3210 trouble for malformed header lines. I hope it is now all correct.
3212 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
3213 with the "reply" argument non-NULL. The code, however (which originally
3214 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
3215 but it didn't always do it. This confused somebody who was copying the
3216 code for some other use. I have removed all the tests.
3218 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
3219 feature that was used to support insecure browsers during the U.S. crypto
3220 embargo. It requires special client support, and Exim is probably the
3221 only MTA that supported it -- and would never use it because real RSA is
3222 always available. This code has been removed, because it had the bad
3223 effect of slowing Exim down by computing (never used) parameters for the
3224 RSA_EXPORT functionality.
3226 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
3227 authenticator to fail if there's a tab character in the incoming data
3228 (there should never be unless someone is messing about, as it's supposed
3229 to be base64-encoded). Also added, on Timo's advice, the "secured" option
3230 if the connection is using TLS or if the remote IP is the same as the
3231 local IP, and the "valid-client-cert option" if a client certificate has
3234 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
3235 authenticators. This can be used for authorization after authentication
3236 succeeds. (In the case of plaintext, it servers for both authentication
3239 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
3240 if any retry times were supplied.
3242 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
3243 connection, where there is no client IP address to check. In this
3244 situation, the verify now always succeeds.
3246 PH/25 Applied John Jetmore's -Mset patch.
3248 PH/26 Added -bem to be like -Mset, but loading a message from a file.
3250 PH/27 In a string expansion for a processed (not raw) header when multiple
3251 headers of the same name were present, leading whitespace was being
3252 removed from all of them, but trailing whitespace was being removed only
3253 from the last one. Now trailing whitespace is removed from each header
3254 before concatenation. Completely empty headers in a concatenation (as
3255 before) are ignored.
3257 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
3258 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
3260 PH/29 [Removed. This was a change that I later backed out, and forgot to
3261 correct the ChangeLog entry (that I had efficiently created) before
3262 committing the later change.]
3264 PH/30 Exim was sometimes attempting to deliver messages that had suffered
3265 address errors (4xx response to RCPT) over the same connection as other
3266 messages routed to the same hosts. Such deliveries are always "forced",
3267 so retry times are not inspected. This resulted in far too many retries
3268 for the affected addresses. The effect occurred only when there were more
3269 hosts than the hosts_max_try setting in the smtp transport when it had
3270 the 4xx errors. Those hosts that it had tried were not added to the list
3271 of hosts for which the message was waiting, so if all were tried, there
3272 was no problem. Two fixes have been applied:
3274 (i) If there are any address or message errors in an SMTP delivery, none
3275 of the hosts (tried or untried) are now added to the list of hosts
3276 for which the message is waiting, so the message should not be a
3277 candidate for sending over the same connection that was used for a
3278 successful delivery of some other message. This seems entirely
3279 reasonable: after all the message is NOT "waiting for some host".
3280 This is so "obvious" that I'm not sure why it wasn't done
3281 previously. Hope I haven't missed anything, but it can't do any
3282 harm, as the worst effect is to miss an optimization.
3284 (ii) If, despite (i), such a delivery is accidentally attempted, the
3285 routing retry time is respected, so at least it doesn't keep
3286 hammering the server.
3288 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
3289 in ${readsocket because some servers need this prod.
3291 PH/32 Added some extra debug output when updating a wait-xxx database.
3293 PH/33 The hint "could be header name not terminated by colon", which has been
3294 given for certain expansion errors for a long time, was not being given
3295 for the ${if def:h_colon_omitted{... case.
3297 PH/34 The spec says: "With one important exception, whenever a domain list is
3298 being scanned, $domain contains the subject domain." There was at least
3299 one case where this was not true.
3301 PH/35 The error "getsockname() failed: connection reset by peer" was being
3302 written to the panic log as well as the main log, but it isn't really
3303 panic-worthy as it just means the connection died rather early on. I have
3304 removed the panic log writing for the ECONNRESET error when getsockname()
3307 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
3308 runs only) independently of the message's sender address. This meant
3309 that, if the 4xx error was in fact related to the sender, a different
3310 message to the same recipient with a different sender could confuse
3311 things. In particular, this can happen when sending to a greylisting
3312 server, but other circumstances could also provoke similar problems.
3313 I have changed the default so that the retry time for these errors is now
3314 based a combination of the sender and recipient addresses. This change
3315 can be overridden by setting address_retry_include_sender=false in the
3318 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
3319 remote server are returned as part of bounce messages. This was not
3320 happening for LMTP over a pipe (the lmtp transport), but now it is the
3321 same for both kinds of LMTP.
3323 PH/38 Despite being documented as not happening, Exim was rewriting addresses
3324 in header lines that were in fact CNAMEs. This is no longer the case.
3326 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
3327 and queue runs started by the daemon processed all messages. This has
3328 been fixed so that -R and -S can now usefully be given with -q<time>.
3330 PH/40 Import PCRE release 6.7 (fixes some bugs).
3332 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
3334 PH/42 Give an error if -q is specified more than once.
3336 PH/43 Renamed the variables $interface_address and $interface_port as
3337 $received_ip_address and $received_port, to make it clear that these
3338 values apply to message reception, and not to the outgoing interface when
3339 a message is delivered. (The old names remain recognized, of course.)
3341 PH/44 There was no timeout on the connect() call when using a Unix domain
3342 socket in the ${readsocket expansion. There now is.
3344 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
3345 be meaningful with "accept".
3347 SC/01 Eximstats V1.43
3348 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
3350 SC/02 Eximstats V1.44
3351 Use a glob alias rather than an array ref in the generated
3352 parser. This improves both readability and performance.
3354 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
3355 Collect SpamAssassin and rejection statistics.
3356 Don't display local sender or destination tables unless
3357 there is data to show.
3358 Added average volumes into the top table text output.
3360 SC/04 Eximstats V1.46
3361 Collect data on the number of addresses (recipients)
3362 as well as the number of messages.
3364 SC/05 Eximstats V1.47
3365 Added 'Message too big' to the list of mail rejection
3366 reasons (thanks to Marco Gaiarin).
3368 SC/06 Eximstats V1.48
3369 Mainlog lines which have GMT offsets and are too short to
3370 have a flag are now skipped.
3372 SC/07 Eximstats V1.49 (Alain Williams)
3373 Added the -emptyok flag.
3375 SC/08 Eximstats V1.50
3376 Fixes for obtaining the IP address from reject messages.
3378 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
3379 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
3380 whitespace changes from 4.64-PH/27
3382 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
3385 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
3386 are found, allow negative numbers in numeric criteria)
3388 JJ/06 exipick.20061117.2, added new $message_body_missing variable
3390 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
3391 to match changes made in 4.64-PH/43
3393 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
3395 PH/47 Put in an explicit test for a DNS lookup of an address record where the
3396 "domain" is actually an IP address, and force a failure. This locks out
3397 those revolvers/nameservers that support "A-for-A" lookups, in
3398 contravention of the specifications.
3400 PH/48 When a host name was looked up from an IP address, and the subsequent
3401 forward lookup of the name timed out, the host name was left in
3402 $sender_host_name, contrary to the specification.
3404 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
3405 restricted to single-key lookups, Exim was not diagnosing an error if
3406 * or *@ was used with a query-style lookup.
3408 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
3410 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
3411 long ago, but noone interested enough thought of it. Let's just say that
3412 the "1.1" means that there are some new functions that weren't there at
3413 some point in the past.
3415 PH/51 Error processing for expansion failure of helo_data from an smtp
3416 transport during callout processing was broken.
3418 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
3419 tested/used via the -bh/-bhc/-bs options.
3421 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
3422 bug, fixed in subsequent PCRE releases).
3424 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
3425 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
3427 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
3433 SC/01 Use a glob alias rather than an array ref in eximstats generated
3434 parser. This improves both readability and performance.
3436 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
3437 Don't display local sender or destination tables in eximstats unless
3438 there is data to show.
3439 Added average volumes into the eximstats top table text output.
3441 SC/03 Collect data on the number of addresses (recipients) as well
3442 as the number of messages in eximstats.
3444 TF/01 Correct an error in the documentation for the redirect router. Exim
3445 does (usually) call initgroups() when daemonizing.
3447 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
3448 with consistent privilege compared to when running as a daemon.
3450 TF/03 Note in the spec that $authenticated_id is not set for local
3451 submissions from trusted users.
3453 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
3454 Thanks to Dean Brooks <dean@iglou.com> for the patch.
3456 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
3457 by adding some example configuration directives to the default
3458 configuration file. A little bit of work is required to uncomment the
3459 directives and define how usernames and passwords are checked, but
3460 there is now a framework to start from.
3462 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
3463 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
3464 without this. I don't know how relevant this is to other LDAP libraries.
3466 PH/02 Add the verb name to the "unknown ACL verb" error.
3468 PH/03 Magnus Holmgren's patch for filter_prepend_home.
3470 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
3472 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
3473 directory not expanded when it should be if an expanded home directory
3474 was set for the address (which is overridden by the transport).
3476 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3479 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3480 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3481 because it is too late at that time, and has no effect.
3483 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3484 security issue with \' (bugzilla #107). I could not use the
3485 PQescapeStringConn() function, because it needs a PGconn value as one of
3488 PH/08 When testing addresses using -bt, indicate those final addresses that
3489 are duplicates that would not cause an additional delivery. At least one
3490 person was confused, thinking that -bt output corresponded to deliveries.
3491 (Suppressing duplicates isn't a good idea as you lose the information
3492 about possibly different redirections that led to the duplicates.)
3494 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3495 systems where poll() doesn't work, in particular OS X.
3497 PH/10 Added more information to debugging output for retry time not reached.
3499 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3500 operations in malware.c.
3502 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3505 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3506 syslog_duplication set false, log lines that would normally be written
3507 both the the main log and to the reject log were not written to syslog at
3510 PH/14 In the default configuration, change the use of "message" in ACL warn
3511 statements to "add_header".
3513 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3514 not followed by a command (e.g. "seen endif").
3516 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3517 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3520 PH/17 Added extra conditions to the default value of delay_warning_condition
3524 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3525 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3526 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3529 The Auto-Submitted: and various List- headers are standardised, whereas I
3530 don't think Precedence: ever was.
3532 PH/18 Refactored debugging code in route_finduser() to show more information,
3533 in particular, the error code if getpwnam() issues one.
3535 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3536 This is apparently needed in addition to the PH/07 change above to avoid
3537 any possible encoding problems.
3539 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3540 but not after initializing Perl.
3542 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3543 output them only if debugging. By default they are written stderr,
3544 apparently, which is not desirable.
3546 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3549 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3552 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3554 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3555 authenticated or an ident call has been made. Suppress the default
3556 values for $authenticated_id and $authenticated_sender (but permit -oMai
3557 and -oMas) when testing with -bh.
3559 PH/24 Re-jigged the order of the tests in the default configuration so that the
3560 tests for valid domains and recipients precede the DNS black list and CSA
3561 tests, on the grounds that those ones are more expensive.
3563 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3564 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3565 This bug exists in every version of Exim that I still have, right back to
3568 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3569 However, an attempt to turn on case-sensitivity in a regex key by
3570 including (?-i) didn't work because the subject string was already
3571 lowercased, and the effects were non-intuitive. It turns out that a
3572 one-line patch can be used to allow (?-i) to work as expected.
3578 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3579 other effects) broke the use of negated acl sub-conditions.
3581 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3584 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3585 "Deny" causes Exim to reject the incoming connection with a 554 error.
3586 Unfortunately, if there is a major crisis, such as a disk failure,
3587 tcp-wrappers gives "deny", whereas what one would like would be some
3588 kind of temporary error. A kludge has been added to help with this.
3589 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3590 554 error is used if errno is still zero or contains ENOENT (which occurs
3591 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3594 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3596 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3597 errors. Otherwise a message that provokes a temporary error (when other
3598 messages do not) can cause a whole host to time out.
3600 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3601 addresses were routed directly to files or pipes from a redirect router.
3602 File deliveries just didn't batch; pipe deliveries might have suffered
3605 PH/06 A failure to get a lock for a hints database would erroneously always say
3606 "Failed to get write lock", even when it was really a read lock.
3608 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3609 of 0600. This has been changed to use the value of the lockfile_mode
3610 option (which defaults to 0600).
3612 PH/08 Applied small patch from the Sieve maintainer.
3614 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3615 folder from quota calculations, a direct delivery into this folder messed
3616 up the contents of the maildirsize file. This was because the regex was
3617 used only to exclude .Trash (or whatever) when the size of the mailbox
3618 was calculated. There was no check that a delivery was happening into an
3619 excluded directory. This bug has been fixed by ignoring all quota
3620 processing for deliveries into excluded directories.
3622 PH/10 Added the maildirfolder_create_regex option to appendfile.
3628 PH/01 The code for finding all the local interface addresses on a FreeBSD
3629 system running IPv6 was broken. This may well have applied to all BSD
3630 systems, as well as to others that have similar system calls. The broken
3631 code found IPv4 interfaces correctly, but gave incorrect values for the
3632 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3633 that it would not match correctly against @[] and not recognize the IPv6
3636 PH/02 The ipliteral router was not recognizing addresses of the form user@
3637 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3639 PH/03 Added disable_ipv6.
3641 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3642 decoded form, because it is most often used to construct To: headers
3643 lines in autoreplies, and the decoded form may well be syntactically
3644 invalid. However, $reply_address has leading white space removed, and all
3645 newlines turned into spaces so that the autoreply transport does not
3648 PH/05 If group was specified without a user on a router, and no group or user
3649 was specified on a transport, the group from the router was ignored.
3651 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3652 for visible compile-time settings that can be used to change these
3653 numbers, for those that want even more. Backwards compatibility with old
3654 spool files has been maintained. However, going back to a previous Exim
3655 release will lost any variables that are in spool files.
3657 PH/07 Two small changes when running in the test harness: increase delay when
3658 passing a TCP/IP connection to a new process, in case the original
3659 process has to generate a bounce, and remove special handling of
3660 127.0.0.2 (sic), which is no longer necessary.
3662 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3663 be the same on different OS.
3665 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3668 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3669 whether --show-vars was specified or not
3671 JJ/02 exipick: Added support for new ACL variable spool format introduced
3674 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3675 syntactically invalid From: or Reply-to: line, and a filter used this to
3676 generate an autoreply, and therefore failed to obtain an address for the
3677 autoreply, Exim could try to deliver to a non-existent relative file
3678 name, causing unrelated and misleading errors. What now happens is that
3679 it logs this as a hard delivery error, but does not attempt to create a
3682 PH/11 The exinext utility has a -C option for testing purposes, but although
3683 the given file was scanned by exinext itself; it wasn't being passed on
3684 when Exim was called.
3686 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3687 an end-of-file indication when reading a command response.
3689 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3690 compiled. In many other places in Exim, IPv6 addresses are always
3691 recognized, so I have changed this. It also means that IPv4 domain
3692 literals of the form [IPV4:n.n.n.n] are now always recognized.
3694 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3695 used if the router is not running as root, for example, when verifying at
3696 ACL time, or when using -bh. The debugging output from this situation was
3697 non-existent - all you got was a failure to exec. I have made two
3700 (a) Failures to set uid/gid, the current directory, or a process leader
3701 in a subprocess such as that created by queryprogram now generate
3702 suitable debugging output when -d is set.
3704 (b) The queryprogram router detects when it is not running as root,
3705 outputs suitable debugging information if -d is set, and then runs
3706 the subprocess without attempting to change uid/gid.
3708 PH/15 Minor change to Makefile for building test_host (undocumented testing
3711 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3712 additional section of a DNS packet that returns MX or SRV records.
3713 Instead, it always explicitly searches for A/AAAA records. This avoids
3714 major problems that occur when a DNS server includes only records of one
3715 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3716 fixed another bug: if SRV records were looked up and the corresponding
3717 address records were *not* found in the additional section, the port
3718 values from the SRV records were lost.
3720 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3721 using the correct key (the original address) when searching the retry
3722 rules in order to find which one to use for generating the retry hint.
3724 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3725 adding the default one. Similarly, if it contains a Reply-To: header, the
3726 errors_reply_to option, if set, is not used.
3728 PH/19 When calculating a retry time, Exim used to measure the "time since
3729 failure" by looking at the "first failed" field in the retry record. Now
3730 it does not use this if it is later than than the arrival time of the
3731 message. Instead it uses the arrival time. This makes for better
3732 behaviour in cases where some deliveries succeed, thus re-setting the
3733 "first failed" field. An example is a quota failure for a huge message
3734 when small messages continue to be delivered. Without this change, the
3735 "time since failure" will always be short, possible causing more frequent
3736 delivery attempts for the huge message than are intended.
3737 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3739 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3740 $1, $2, $3) because the numerical variables can be reset during some
3741 expansion items (e.g. "match"), thereby losing the authentication data.
3743 PH/21 Make -bV show the size of off_t variables so that the test suite can
3744 decide whether to run tests for quotas > 2G.
3746 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3747 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3748 filecount value is greater than 2G or if a quota value is greater than 2G
3749 on a system where the size of off_t is not greater than 4, a panic error
3752 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3753 never match. The debug and -bh output now contains an explicit error
3754 message indicating a malformed IPv4 address or mask.
3756 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3757 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3758 PH/23 above applies.
3760 PH/25 Do not write to syslog when running in the test harness. The only
3761 occasion when this arises is a failure to open the main or panic logs
3762 (for which there is an explicit test).
3764 PH/26 Added the /no_tell option to "control=freeze".
3766 PH/27 If a host name lookup failed very early in a connection, for example, if
3767 the IP address matched host_lookup and the reverse lookup yielded a name
3768 that did not have a forward lookup, an error message of the form "no IP
3769 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3770 could be logged. Now it outputs the IP address instead of "NULL".
3772 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3773 allows the sender and the authenticated sender to be set when
3774 submitting a message from within Exim. Since child_open_exim() is
3775 documented for local_scan(), the new function should be too.
3777 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3778 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3779 results in an empty string is now treated as unset.
3781 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3783 PH/31 Added sender_verify_fail logging option.
3785 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3786 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3787 tidied the source and removed it altogether.
3789 PH/33 When a queue run was abandoned because the load average was too high, a
3790 log line was always written; now it is written only if the queue_run log
3791 selector is set. In addition, the log line for abandonment now contains
3792 information about the queue run such as the pid. This is always present
3793 in "start" and "stop" lines but was omitted from the "abandon" line.
3795 PH/34 Omit spaces between a header name and the colon in the error message that
3796 is given when verify = headers_syntax fails (if there are lots of them,
3797 the message gets confusing).
3799 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3800 names, as there are now some PTR records that contain slashes. This check
3801 is only to protect against broken name servers that fall over on strange
3802 characters, so the fact that it applies to all lookups doesn't matter.
3804 PH/36 Now that the new test suite is complete, we can remove some of the
3805 special code in Exim that was needed for the old test suite. For example,
3806 sorting DNS records because real resolvers return them in an arbitrary
3807 order. The new test suite's fake resolver always returns records in the
3810 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3811 bounces) except when queue_only is set, to avoid logging races between
3812 the different processes.
3814 PH/38 Panic-die if .include specifies a non-absolute path.
3816 PH/39 A tweak to the "H" retry rule from its user.
3818 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3819 a label. They prevented compilation on older perls.
3821 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3822 a warning to be raised on newish perls.
3824 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3825 on queue. Changes to match documented behaviour of showing count of
3826 messages matching specified criteria.
3828 PH/40 Changed the default ident timeout from 30s to 5s.
3830 PH/41 Added support for the use of login_cap features, on those BSD systems
3831 that have them, for controlling the resources used by pipe deliveries.
3833 PH/42 The content-scanning code uses fopen() to create files in which to put
3834 message data. Previously it was not paying any attention to the mode of
3835 the files. Exim runs with umask(0) because the rest of the code creates
3836 files with open(), and sets the required mode explicitly. Thus, these
3837 files were ending up world-writeable. This was not a big issue, because,
3838 being within the spool directory, they were not world-accessible. I have
3839 created a function called modefopen, which takes an additional mode
3840 argument. It sets umask(777), creates the file, chmods it to the required
3841 mode, then resets the umask. All the relevant calls to fopen() in the
3842 content scanning code have been changed to use this function.
3844 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3845 to 24 hours. This avoids potential overflow problems when processing G
3846 and H retry rules. I suspect nobody ever tinkers with this value.
3848 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3850 PH/45 When the plaintext authenticator is running as a client, the server's
3851 challenges are checked to ensure they are valid base64 strings. By
3852 default, the authentication attempt is cancelled if an invalid string is
3853 received. Setting client_ignore_invalid_base64 true ignores these errors.
3854 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3855 they are received. Thus, the responses can be made to depend on the
3856 challenges. If an invalid string is ignored, an empty string is placed in
3859 PH/46 Messages that are created by the autoreply transport now contains a
3860 References: header, in accordance with RFCs 2822 and 3834.
3862 PH/47 Added authenticated_sender_force to the smtp transport.
3864 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3866 PH/49 Installed latest patch from the Sieve maintainer.
3868 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3869 was also set, the appendfile transport was unnecessarily scanning a
3870 directory of message files (e.g. for maildir delivery) to find the count
3871 of files (along with the size), even though it did not need this
3872 information. It now does the scan only if it needs to find either the
3873 size of the count of files.
3875 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3877 PH/52 Two bugs concerned with error handling when the smtp transport is
3880 (i) Exim was not creating retry information for temporary errors given
3881 for individual recipients after the DATA command when the smtp transport
3882 was used in LMTP mode. This meant that they could be retried too
3883 frequently, and not timed out correctly.
3885 (ii) Exim was setting the flag that allows error details to be returned
3886 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3887 recipients that were returned after the DATA command.
3889 PH/53 This is related to PH/52, but is more general: for any failing address,
3890 when detailed error information was permitted to be returned to the
3891 sender, but the error was temporary, then after the final timeout, only
3892 "retry timeout exceeded" was returned. Now it returns the full error as
3893 well as "retry timeout exceeded".
3895 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3896 do this, and (what is worse) MTAs that accept it.
3898 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3899 will now be deprecated.
3901 PH/56 New os.c-cygwin from the Cygwin maintainer.
3903 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3904 formats (previously only available in exim formats via -bpr, -bpru,
3905 and -bpra. Now also available in native and exiqgrep formats)
3907 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3908 with very large, slow to parse queues
3910 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3912 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3914 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3915 responses to authentication challenges, though it was showing the
3916 challenges; (ii) I've removed the CR characters from the debug output for
3919 PH/58 Allow for the insertion of a newline as well as a space when a string
3920 is turned into more than one encoded-word during RFC 2047 encoding. The
3921 Sieve code now uses this.
3923 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3924 data_4xx, lost_connection, tls_required.
3926 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3927 message was being sent as an SMTP response.
3929 PH/61 Add -l and -k options to exicyclog.
3931 PH/62 When verifying, if an address was redirected to one new address, so that
3932 verification continued, and the new address failed or deferred after
3933 having set something in $address_data, the value of $address_data was not
3934 passed back to the ACL. This was different to the case when no
3935 redirection occurred. The value is now passed back in both cases.
3937 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3938 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3939 use login_cap.h, so on its own it isn't the distinguishing feature. The
3940 new name refers directly to the setclassresources() function.
3942 PH/65 Added configuration files for NetBSD3.
3944 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3946 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3947 is preferred over IPv4.
3949 PH/68 The bounce_return_message and bounce_return_body options were not being
3950 honoured for bounces generated during the reception of non-SMTP messages.
3951 In particular, this applied to messages rejected by the ACL. This bug has
3952 been fixed. However, if bounce_return_message is true and bounce_return_
3953 body is false, the headers that are returned for a non-SMTP message
3954 include only those that have been read before the error was detected.
3955 (In the case of an ACL rejection, they have all been read.)
3957 PH/69 The HTML version of the specification is now built in a directory called
3958 spec_html instead of spec.html, because the latter looks like a path with
3959 a MIME-type, and this confuses some software.
3961 PH/70 Catch two compiler warnings in sieve.c.
3963 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3964 function verify_get_ident() calls ip_connect() to connect a socket, but
3965 if the "connect()" function timed out, ip_connect() used to close the
3966 socket. However, verify_get_ident() also closes the socket later, and in
3967 between Exim writes to the log, which may get opened at this point. When
3968 the socket was closed in ip_connect(), the log could get the same file
3969 descriptor number as the socket. This naturally causes chaos. The fix is
3970 not to close the socket in ip_connect(); the socket should be closed by
3971 the function that creates it. There was only one place in the code where
3972 this was missing, in the iplookup router, which I don't think anybody now
3973 uses, but I've fixed it anyway.
3975 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3976 well as to direct DNS lookups. Otherwise the handling of names in host
3977 lists is inconsistent and therefore confusing.
3983 PH/01 Two changes to the default runtime configuration:
3985 (1) Move the checks for relay_from_hosts and authenticated clients from
3986 after to before the (commented out) DNS black list checks.
3988 (2) Add control=submission to the relay_from_hosts and authenticated
3989 clients checks, on the grounds that messages accepted by these
3990 statements are most likely to be submissions.
3992 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3994 (1) Generate an error if the third argument for the ${prvs expansion is
3997 (2) Treat a missing third argument of ${prvscheck as if it were an empty
4000 (3) Reset the variables that are obtained from the first argument of
4001 ${prvscheck and used in the second argument before leaving the code,
4002 because their memory is reclaimed, so using them afterwards may do
4005 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
4006 one (it's much easier than Tom thought :-).
4008 (5) Because of (4), we can now allow for the use of $prvscheck_result
4009 inside the third argument.
4011 PH/03 For some reason, the default setting of PATH when running a command from
4012 a pipe transport was just "/usr/bin". I have changed it to
4015 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
4016 anything to be listed in the output from -bV.
4018 PH/05 When a filter generated an autoreply, the entire To: header line was
4019 quoted in the delivery log line, like this:
4021 => >A.N.Other <ano@some.domain> <original@ddress> ...
4023 This has been changed so that it extracts the operative address. There
4024 may be more than one such address. If so, they are comma-separated, like
4027 => >ano@some.domain,ona@other.domain <original@ddress> ...
4029 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
4030 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
4031 not being looked up in the rDNS to get a host name, Exim was showing the
4032 IP address twice in Received: lines, even though the IP addresses were
4033 identical. For example:
4035 Received: from [1.2.3.4] (helo=[1.2.3.4])
4037 However, if the real host name was known, it was omitting the HELO data
4038 if it matched the actual IP address. This has been tidied up so that it
4039 doesn't show the same IP address twice.
4041 PH/07 When both +timestamp and +memory debugging was on, the value given by
4042 $tod_xxx expansions could be wrong, because the tod_stamp() function was
4043 called by the debug printing, thereby overwriting the timestamp buffer.
4044 Debugging no longer uses the tod_stamp() function when +timestamp is set.
4046 PH/08 When the original message was included in an autoreply transport, it
4047 always said "this is a copy of the message, including all the headers",
4048 even if body_only or headers_only was set. It now gives an appropriate
4051 PH/09 Applied a patch from the Sieve maintainer which:
4053 o fixes some comments
4054 o adds the (disabled) notify extension core
4055 o adds some debug output for the result of if/elsif tests
4056 o points to the current vacation draft in the documentation
4057 and documents the missing references header update
4061 o fixes a bug in processing the envelope test (when testing
4062 multiple envelope elements, the last element determined the
4065 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
4066 Electronic Mail") by including:
4068 Auto-submitted: auto-generated
4070 in the messages that it generates (bounce messages and others, such as
4071 warnings). In the case of bounce messages for non-SMTP messages, there was
4072 also a typo: it was using "Auto_submitted" (underscore instead of
4073 hyphen). Since every message generated by Exim is necessarily in response
4074 to another message, thes have all been changed to:
4076 Auto-Submitted: auto-replied
4078 in accordance with these statements in the RFC:
4080 The auto-replied keyword:
4082 - SHOULD be used on messages sent in direct response to another
4083 message by an automatic process,
4085 - MUST NOT be used on manually-generated messages,
4087 - MAY be used on Delivery Status Notifications (DSNs) and Message
4088 Disposition Notifications (MDNs),
4090 - MUST NOT be used on messages generated by automatic or periodic
4091 processes, except for messages which are automatic responses to
4094 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
4095 to the default Received: header definition.
4097 PH/12 Added log selector acl_warn_skipped (default on).
4099 PH/13 After a successful wildlsearch lookup, discard the values of numeric
4100 variables because (a) they are in the wrong storage pool and (b) even if
4101 they were copied, it wouldn't work properly because of the caching.
4103 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
4104 checking when decoding. Apparently there are clients that generate
4105 overlong encoded strings. Why am I not surprised?
4107 PH/15 If the first argument of "${if match_address" was not empty, but did not
4108 contain an "@" character, Exim crashed. Now it writes a panic log message
4109 and treats the condition as false.
4111 PH/16 In autoreply, treat an empty string for "once" the same as unset.
4113 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
4114 extension "envelope-auth". The code is finished and in agreement with
4115 other implementations, but there is no documentation so far and in fact,
4116 nobody wrote the draft yet. This extension is currently #undef'ed, thus
4117 not changing the active code.
4119 Print executed "if" and "elsif" statements when debugging is used. This
4120 helps a great deal to understand what a filter does.
4122 Document more things not specified clearly in RFC3028. I had all this
4123 sorted out, when out of a sudden new issues came to my mind. Oops."
4125 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
4128 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
4129 canonical form (as documented). However, after a host name lookup from
4130 the IP address, check_host() was doing a simple string comparison with
4131 addresses acquired from the DNS when checking that the found name did
4132 have the original IP as one of its addresses. Since any found IPv6
4133 addresses are likely to be in abbreviated form, the comparison could
4134 fail. Luckily, there already exists a function for doing the comparison
4135 by converting both addresses to binary, so now that is used instead of
4136 the text comparison.
4138 PH/20 There was another similar case to PH/19, when a complete host name was
4139 given in a host list; looking up its IP address could give an abbreviated
4140 form, whereas the current host's name might or might not be abbreviated.
4141 The same fix has been applied.
4147 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
4148 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
4151 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
4152 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
4154 PH/03 Typo: missing ".o" in src/pcre/Makefile.
4156 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
4157 header line, restrict the check to what is listed in RFCs 2369 and 2929.
4158 Also, for "Auto-Submitted", treat anything other than "no" as
4159 non-personal, in accordance with RFC 3834. (Previously it treated
4160 anything starting "auto-" as non-personal.)
4162 TF/01 The control=submission/name=... option had a problem with syntax
4163 errors if the name included a slash character. The /name= option
4164 now slurps the rest of the string, so it can include any characters
4165 but it must come last in the list of options (after /sender_retain
4168 PH/05 Some modifications to the interface to the fake nameserver for the new
4176 TK/01 Added the "success_on_redirect" address verification option. See
4177 NewStuff for rationale and an example.
4179 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
4181 PH/02 Patch to exigrep to allow it to work on syslog lines.
4183 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
4184 fread() to skip over the body file's header line, because in Cygwin the
4185 header line is locked and is inaccessible.
4187 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
4188 co-exist for some time) to make it clear that it is the Exim ID that is
4189 referenced, not the Message-ID: header line.
4191 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
4192 string_format() function, because snprintf() does not exist on all
4195 PH/06 The use of forbid_filter_existstest now also locks out the use of the
4196 ${stat: expansion item.
4198 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
4199 protocol synchronization error", to keep the pedants happy.
4201 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
4202 well as for IRIX systems, when gcc is being used. See the host.c source
4205 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
4207 PH/10 Named domain lists were not working if used in a queue_smtp_domains
4210 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
4211 transport and to the smtp transport in LMTP mode.
4213 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
4215 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
4216 run a filter in a subprocess. This could lead to confusion in subsequent
4217 lookups in the parent process. There should also be a search_tidyup() at
4218 the end of the subprocess.
4220 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
4221 only if the host matched helo_try_verify_hosts, which caused the
4222 verification to occur when the EHLO/HELO command was issued. The ACL just
4223 tested the remembered result. Now, if a previous verification attempt has
4224 not happened, "verify = helo" does it there and then.
4226 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
4228 TK/03 Fix log output including CR from clamd.
4230 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
4231 exist provoked a memory error which could cause a segfault.
4233 PH/15 Installed PCRE 6.2
4235 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
4237 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
4238 of the problem. Specifically, suggested +O2 rather than +O1 for the
4241 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
4243 PH/20 If a delivery was routed to a non-standard port by means of an SRV
4244 record, the port was not correctly logged when the outgoing_port log
4245 selector was set (it logged the transort's default port).
4247 PH/21 Added support for host-specific ports to manualroute, queryprogram,
4248 fallback_hosts, and "hosts" in the smtp transport.
4250 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
4251 host errors such as "Connection refused".
4253 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
4254 authentication with radiusclient 0.4.9:
4256 - Error returned from rc_read_config was caught wrongly
4257 - Username/password not passed on to radius server due to wrong length.
4259 The presumption is that some radiusclient API changes for 4.51/PH/17
4260 were not taken care of correctly. The code is still untested by me (my
4261 Linux distribution still has 0.3.2 of radiusclient), but it was
4262 contributed by a Radius user.
4264 PH/24 When doing a callout, the value of $domain wasn't set correctly when
4265 expanding the "port" option of the smtp transport.
4267 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
4268 while reading a MIME header. Thanks to Tom Hughes for a patch.
4270 PH/24 Include config.h inside local_scan.h so that configuration settings are
4273 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
4274 that in an ACL for RCPT (for example), you can examine exactly what was
4277 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
4278 commands, but it was not correctly comparing the address with the actual
4279 client host address. Thus, it would show the EHLO address in Received:
4280 header lines when this was not necessary.
4282 PH/27 Added the % operator to ${eval:}.
4284 PH/28 Exim tries to create and chdir to its spool directory when it starts;
4285 it should be ignoring failures (because with -C, for example, it has lost
4286 privilege). It wasn't ignoring creation failures other than "already
4289 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
4292 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
4293 because an input line was too long, either on its own, or by virtue of
4294 too many continuations, the temporary file was not being removed, and the
4295 return code was incorrect.
4297 PH/31 Missing "BOOL" in function definition in filtertest.c.
4299 PH/32 Applied Sieve patches from the maintainer.
4301 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
4303 PH/33 Added "verify = not_blind".
4305 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
4306 Local/Makefile (with some defaults set). These are used in built scripts
4307 such as exicyclog, but they have never been used in the exim_install
4308 script (though there are many overriding facilities there). I have
4309 arranged that the exim_install script now takes note of these two
4312 PH/35 Installed configuration files for Dragonfly.
4314 PH/36 When a locally submitted message by a trusted user did not contain a
4315 From: header, and the sender address was obtained from -f or from an SMTP
4316 MAIL command, and the trusted user did not use -F to supply a sender
4317 name, $originator_name was incorrectly used when constructing a From:
4318 header. Furthermore, $originator_name was used for submission mode
4319 messages from external hosts without From: headers in a similar way,
4320 which is clearly wrong.
4322 PH/37 Added control=suppress_local_fixups.
4324 PH/38 When log_selector = +received_sender was set, and the addition of the
4325 sender made the log line's construction buffer exactly full, or one byte
4326 less than full, an overflow happened when the terminating "\n" was
4329 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
4330 when the result of a list match is failure because a DNS lookup failed.
4332 PH/40 RM_COMMAND is now used in the building process.
4334 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
4335 the "build-* directories that it finds.
4337 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
4338 address in a domain literal was a prefix of an interface address.
4340 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
4341 when verifying a sender address, unless rewrite_headers is false.
4343 PH/44 Wrote a long comment about why errors_to addresses are verified as
4344 recipients, not senders.
4346 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
4347 the ratelimit ACL was added.
4349 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
4351 PH/46 Added extra information about PostgreSQL errors to the error string.
4353 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
4354 avoiding the need to install special zones in a real server. This is
4355 backwards compatible; if it can't find the fake resolver, it drops back.
4356 Thus, both old and new test suites can be run.
4358 TF/02 Added util/ratelimit.pl
4360 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
4361 clock is set back in time.
4363 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
4364 Candler <B.Candler@pobox.com>.
4366 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
4367 OK for addresses that are the result of redirections.
4369 PH/48 A number of further additions for the benefit of the new test suite,
4370 including a fake gethostbyname() that interfaces to the fake DNS resolver
4373 TF/06 The fix for widen_domains has also been applied to qualify_single and
4374 search_parents which are the other dnslookup options that can cause
4377 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
4380 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
4382 TF/07 Exim produced the error message "an SRV record indicated no SMTP
4383 service" if it encountered an MX record with an empty target hostname.
4384 The message is now "an MX or SRV record indicated no SMTP service".
4386 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
4387 if the DNS of the sending site is misconfigured. This is quite a
4388 common situation. This change restores the behaviour of treating a
4389 helo verification defer as a failure.
4391 PH/51 If self=fail was set on a router, the bounce message did not include the
4392 actual error message.
4398 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
4400 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
4401 command itself ended in error, the underlying message about the transport
4402 filter timeout was being overwritten with the pipe command error. Now the
4403 underlying error message should be appended to the second error message.
4405 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
4407 PH/02 Reduce the amount of output that "make" produces by default. Full output
4408 can still be requested.
4410 PH/03 The warning log line about a condition test deferring for a "warn" verb
4411 was being output only once per connection, rather than after each
4412 occurrence (because it was using the same function as for successful
4413 "warn" verbs). This seems wrong, so I have changed it.
4415 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
4416 it should not have, which might have caused a crash in the right
4417 circumstances, but probably never did.
4419 PH/04 Installed a modified version of Tony Finch's patch to make submission
4420 mode fix the return path as well as the Sender: header line, and to
4421 add a /name= option so that you can make the user's friendly name appear
4424 TF/03 Added the control = fakedefer ACL modifier.
4426 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
4427 Mark Lowes for thorough testing.
4429 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
4431 TK/03 Merged latest SRS patch from Miles Wilton.
4433 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
4434 with the definition in sysexits.h (which is #included earlier).
4435 Fortunately, Exim does not actually use EX_OK. The code used to try to
4436 preserve the sysexits.h value, by assuming that macro definitions were
4437 scanned for macro replacements. I have been disabused of this notion,
4438 so now the code just undefines EX_OK before #including unistd.h.
4440 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
4441 in the smtp transport. When a block could not be written in a single
4442 write() function, the timeout was being re-applied to each part-write.
4443 This seems wrong - if the receiver was accepting one byte at a time it
4444 would take for ever. The timeout is now adjusted when this happens. It
4445 doesn't have to be particularly precise.
4447 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
4448 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
4450 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
4451 without a domain if the check to <postmaster@domain> fails.
4453 SC/01 Eximstats: added -xls and the ability to specify output files
4454 (patch written by Frank Heydlauf).
4456 SC/02 Eximstats: use FileHandles for outputting results.
4458 SC/03 Eximstats: allow any combination of xls, txt, and html output.
4460 SC/04 Eximstats: fixed display of large numbers with -nvr option
4462 SC/05 Eximstats: fixed merging of reports with empty tables.
4464 SC/06 Eximstats: added the -include_original_destination flag
4466 SC/07 Eximstats: removed tabs and trailing whitespace.
4468 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
4470 TK/06 MBOX spool code: Add real "From " MBOX separator line
4471 so the .eml file is really in mbox format (even though
4472 most programs do not really care). Patch from Alex Miller.
4474 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4475 The latter is generated from $received_to and is only set if the
4476 message has one envelope recipient. SA can use these headers,
4477 obviously out-of-the-box. Patch from Alex Miller.
4479 PH/08 The ${def test on a variable was returning false if the variable's
4480 value was "0", contrary to what the specification has always said!
4481 The result should be true unless the variable is empty.
4483 PH/09 The syntax error of a character other than { following "${if
4484 def:variable_name" (after optional whitespace) was not being diagnosed.
4485 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4486 accidental colon was present, for example, could give incorrect results.
4488 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4489 result is used (not including appendfile, where other changes are about
4492 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4493 This involved changing a lot of size variables from int to off_t. It
4494 should work with maildirs and everything.
4496 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4497 spamd dying while we are connected to it.
4499 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4502 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4503 function for BDB 4.3.
4505 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4507 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4508 the sources for PCRE have been re-arranged and more files are now
4511 PH/15 The code I had for printing potentially long long variables in PH/11
4512 above was not the best (it lost precision). The length of off_t variables
4513 is now inspected at build time, and an appropriate printing format (%ld
4514 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4515 to be "long long int" or "long int". This is needed for the internal
4516 formatting function string_vformat().
4518 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4519 the configuration file to be ":syslog", then the script "guesses" where
4520 the logs files are, rather than using the compiled in default. In our
4521 case the guess is not the same as the compiled default, so the script
4522 suddenly stopped working when I started to use syslog. The patch checks
4523 to see if log_file_path is "". If so, it attempts to read it from exim
4524 with no configuration file to get the compiled in version, before it
4525 falls back to the previous guessing code."
4527 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4528 implementing BATV in an Exim configuration. See NewStuff for the gory
4531 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4532 Makefile that are specific to HP-UX.
4534 PH/18 If the "use_postmaster" option was set for a recipient callout together
4535 with the "random" option, the postmaster address was used as the MAIL
4536 FROM address for the random test, but not for the subsequent recipient
4537 test. It is now used for both.
4539 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4540 patch removes a few documentation additions to RFC 3028, because the
4541 latest draft now contains them. It adds the new en;ascii-case comparator
4542 and a new error check for 8bit text in MIME parts. Comparator and
4543 require names are now matched exactly. I enabled the subaddress
4544 extension, but it is not well tested yet (read: it works for me)."
4546 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4547 rework some of the code of TK/09 above to avoid the hardwired use of
4548 "%lld" and "long long". Replaced the call to snprintf() with a call to
4551 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4552 records point to non-existent hosts", "retry timeout exceeded", and
4553 "retry time not reached for any host after a long failure period".
4555 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4556 experimental DomainKeys support:
4558 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4559 (2) On an error such as an illegally used "control", the wrong name for
4560 the control was given.
4562 These problems did NOT occur unless DomainKeys support was compiled.
4564 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4566 PH/24 Added ${if match_ip condition.
4568 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4569 databases so that it will be absolutely obvious if a crash occurs in the
4570 DB library. This is a regular occurrence (often caused by mis-matched
4573 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4574 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4575 picked up on a user's system that detects such things. There doesn't seem
4576 to be a gcc warning option for this - only an attribute that has to be
4577 put on the function's prototype. It seems that in Fedora Core 4 they have
4578 set this on a number of new functions. No doubt there will be more in due
4581 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4582 specify a transport. However, if an address that was verified by such a
4583 router was the subject of a callout, Exim crashed because it tried to
4584 read the rcpt_include_affixes from the non-existent transport. Now it
4585 just assumes that the setting of that option is false. This bug was
4586 introduced by 4.51/PH/31.
4588 PH/28 Changed -d+all to exclude +memory, because that information is very
4589 rarely of interest, but it makes the output a lot bigger. People tend to
4590 do -d+all out of habit.
4592 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4593 code in os-type was giving problems when libc.so lives in lib64, like on
4596 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4597 aren't the modern standard, and it seems that some systems' include files
4598 don't always have them. Exim was already checking for some of the newer
4599 ones like T_AAAA, and defining it itself. I've added checks for all the
4600 record types that Exim uses.
4602 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4603 not automatically generating a new one, as it is supposed to. This
4604 prevented TLS from working. If the file did exist, but contained invalid
4605 data, a new version was generated, as expected. It was only the case of a
4606 non-existent file that was broken.
4608 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4609 with a change in libdomainkeys > 0.64.
4611 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4612 from DNS. If the selector record carries the flag, it now has
4613 precedence over the domain-wide flag.
4615 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4617 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4618 the use of a port name that isn't defined in /etc/services) occurred, the
4619 message was deferred as in a normal delivery, and thus remained on the
4620 spool, instead of being failed because of the mua_wrapper setting. This
4621 is now fixed, and I tidied up some of the mua_wrapper messages at the
4624 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4625 the messages in a hash of arrays rather than using individual hashes.
4626 This is a bit cleaner and results in dramatic memory savings, albeit
4627 at a slight CPU cost.
4629 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4630 as requested by Marc Sherman.
4632 SC/10 Eximstats: added histograms for user specified patterns as requested
4635 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4637 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4638 fopen() in the content-scanning modules that did not already have it.
4644 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4645 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4647 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4649 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4651 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4652 to test Sieve filters that use "vacation".
4654 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4655 that changes the way the GnuTLS parameters are stored in the cache file.
4656 The new format can be generated externally. For backward compatibility,
4657 if the data in the cache doesn't make sense, Exim assumes it has read an
4658 old-format file, and it generates new data and writes a new file. This
4659 means that you can't go back to an older release without removing the
4662 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4663 work if there are any delivery delays because "one_time" forces the
4664 parent to be marked "delivered", so its unseen clone is never tried
4665 again. For this reason, Exim now forbids the simultaneous setting of
4668 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4669 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4670 entry if you want to know the details. The fix, however, neglected to
4671 consider the case when local delivery batching is involved. The test for
4672 "previously delivered" was not happening when checking to see if an
4673 address could be batched with a previous (undelivered) one; under
4674 certain circumstances this could lead to multiple deliveries to the same
4677 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4678 in its include files, and this causes problems building Exim.
4680 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4681 header_syntax) but Exim was just ignoring anything given after a slash.
4682 In particular, this caused confusion with an attempt to use "verify =
4683 reverse_host_lookup/defer_ok". An error is now given when options are
4684 supplied for verify items that do not have them. (Maybe reverse_host_
4685 lookup should have a defer_ok option, but that's a different point.)
4687 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4688 defined by RFC 821) to 2048, because there were problems with some AUTH
4689 commands, and RFC 1869 says the size should be increased for extended
4690 SMTP commands that take arguments.
4692 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4695 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4696 "unknown" error; now it says that the functionality isn't in the binary.
4698 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4699 an address' error message when a string expansion fails (syntax or
4700 whatever). Otherwise the password may appear in the log. Following change
4701 PH/42 below, there is no longer a chance of it appearing in a bounce
4704 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4706 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4707 of its addresses was ever tried. (Bugzilla bug #2.)
4709 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4710 the result incorrectly in the debug output. (It correctly added a newline
4711 to what was transported.)
4713 TF/01 Added $received_time.
4715 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4716 commented out examples of how to interface to a virus scanner and to
4717 SpamAssassin. Also added commented examples of av_scanner and
4718 spamd_address settings.
4720 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4721 and controls are allowed in which ACLs. There were a couple of minor
4722 errors. Some of the entries in the conditions table (which is a table of
4723 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4724 negation of where the condition IS allowed.
4726 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4728 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4729 header file does not have a version number, so I've had to invent a new
4730 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4731 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4732 radiusclient), but it was contributed by a Radius user.
4734 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4735 files or directories, for OpenSSL.
4737 PH/19 When an Exim process that is running as root has to create an Exim log
4738 file, it does so in a subprocess that runs as exim:exim so as to get the
4739 ownership right at creation (otherwise, other Exim processes might see
4740 the file with the wrong ownership). There was no test for failure of this
4741 fork() call, which would lead to the process getting stuck as it waited
4742 for a non-existent subprocess. Forks do occasionally fail when resources
4743 run out. I reviewed all the other calls to fork(); they all seem to check
4746 PH/20 When checking for unexpected SMTP input at connect time (before writing
4747 the banner), Exim was not dealing correctly with a non-positive return
4748 from the read() function. If the client had disconnected by this time,
4749 the result was a log entry for a synchronization error with an empty
4750 string after "input=" when read() returned zero. If read() returned -1
4751 (an event I could not check), uninitialized data bytes were printed.
4752 There were reports of junk text (parts of files, etc) appearing after
4755 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4757 PH/22 Added support for macro redefinition, and (re)definition in between
4758 driver and ACL definitions.
4760 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4761 forgetting to use the resulting value; it was using the unexpanded value.
4763 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4764 hadn't been configured. The fix is from Juergen Kreileder, who
4765 understands it better than I do:
4767 "Here's what I see happening with three configured cyrus_sasl
4768 authenticators configured (plain, login, cram-md5):
4770 On startup auth_cyrus_sasl_init() gets called for each of these.
4771 This means three calls to sasl_listmech() without a specified mech_list.
4772 => SASL tests which mechs of all available mechs actually work
4773 => three warnings about OTP not working
4774 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4776 With the patch, sasl_listmech() also gets called three times. But now
4777 SASL's mech_list option is set to the server_mech specified in the the
4778 authenticator. Or in other words, the answer from sasl_listmech()
4779 gets limited to just the mech you're testing for (which is different
4781 => the return list contains just 'plain' or 'login', 'cram-md5' or
4782 nothing depending on the value of ob->server_mech.
4784 I've just tested the patch: Authentication still works fine,
4785 unavailable mechs specified in the exim configuration are still
4786 caught, and the auth.log warnings about OTP are gone."
4788 PH/25 When debugging is enabled, the contents of the command line are added
4789 to the debugging output, even when log_selector=+arguments is not
4792 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4793 answer is "GNU", and only if the return is "GNU/something" is the answer
4796 PH/27 $acl_verify_message is now set immediately after the failure of a
4797 verification in an ACL, and so is available in subsequent modifiers. In
4798 particular, the message can be preserved by coding like this:
4800 warn !verify = sender
4801 set acl_m0 = $acl_verify_message
4803 Previously, $acl_verify_message was set only while expanding "message"
4804 and "log_message" when a very denied access.
4806 PH/28 Modified OS/os.c-Linux with
4808 -#ifndef OS_LOAD_AVERAGE
4809 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4811 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4812 nomenclature these days.)
4814 PH/29 Installed patch from the Sieve maintainer that adds the options
4815 sieve_useraddress and sieve_subaddress to the redirect router.
4817 PH/30 In these circumstances:
4818 . Two addresses routed to the same list of hosts;
4819 . First host does not offer TLS;
4820 . First host accepts first address;
4821 . First host gives temporary error to second address;
4822 . Second host offers TLS and a TLS session is established;
4823 . Second host accepts second address.
4824 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4825 and peerdn, if requested) that were in fact used only for the second
4828 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4829 attention to any local part prefix or suffix that was matched by the
4830 router that accepted the address. It now behaves in the same way as it
4831 does for delivery: the affixes are removed from the local part unless
4832 rcpt_include_affixes is set on the transport.
4834 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4835 timeout during the DATA phase of an incoming message.
4837 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4838 applied a patch sanctioned by the Sieve maintainer.
4840 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4841 the uid or gid is negative. A case of a negative gid caused this to be
4842 noticed. The fix allows for either to be negative.
4844 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4845 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4848 PH/36 The change PH/12 above was broken. Fixed it.
4850 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4851 the grounds that routing the same address twice would always produce the
4852 same answer. This might have been true once, but it is certainly no
4853 longer true now. Routing a child address may depend on the previous
4854 routing that produced that child. Some complicated redirection strategies
4855 went wrong when messages had multiple recipients, and made Exim's
4856 behaviour dependent on the order in which the addresses were given.
4858 I have moved the duplicate checking until after the routing is complete.
4859 Exim scans the addresses that are assigned to local and remote
4860 transports, and removes any duplicates. This means that more work will be
4861 done, as duplicates will always all be routed, but duplicates are
4862 presumably rare, so I don't expect this is of any significance.
4864 For deliveries to pipes, files, and autoreplies, the duplicate checking
4865 still happens during the routing process, since they are not going to be
4868 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4869 It corrects a timeout issue with spamd. This is Ian's comment: "The
4870 background is that sometimes spamd either never reads data from a
4871 connection it has accepted, or it never writes response data. The exiscan
4872 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4873 blindly assumes that writes won't block so it may never time out."
4875 PH/39 Allow G after quota size as well as K and M.
4877 PH/40 The value set for $authenticated_id in an authenticator may not contain
4878 binary zeroes or newlines because the value is written to log lines and
4879 to spool files. There was no check on this. Now the value is run through
4880 the string_printing() function so that such characters are converted to
4881 printable escape sequences.
4883 PH/41 $message_linecount is a new variable that contains the total number of
4884 lines in the message. Compare $body_linecount, which is the count for the
4887 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4888 bounce and delay warning messages, except in certain special cases, which
4891 (a) An SMTP error message from a remote host;
4892 (b) A message specified in a :fail: redirection;
4893 (c) A message specified in a "fail" command in a system filter;
4894 (d) A message specified in a FAIL return from the queryprogram router;
4895 (e) A message specified by the cannot_route_message router option.
4897 In these cases only, Exim does include the error details in bounce and
4898 warning messages. There are also a few cases where bland messages such
4899 as "unrouteable address" or "local delivery error" are given.
4901 PH/43 $value is now also set for the "else" part of a ${run expansion.
4903 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4904 being worked on, but at least Exim now implements the latest version to
4907 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4908 process to complete was treated as a delivery failure, a timeout while
4909 writing the message to the pipe was logged, but erroneously treated as a
4910 successful delivery. Such timeouts include transport filter timeouts. For
4911 consistency with the overall process timeout, these timeouts are now
4912 treated as errors, giving rise to delivery failures by default. However,
4913 there is now a new Boolean option for the pipe transport called
4914 timeout_defer, which, if set TRUE, converts the failures into defers for
4915 both kinds of timeout. A transport filter timeout is now identified in
4918 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4919 systems where "make" and "gmake" are different, calling "gmake" at top
4920 level broke things. I've arranged for the value of $(MAKE) to be passed
4921 from the Makefile to this script so that it can call the same version of
4925 A note about Exim versions 4.44 and 4.50
4926 ----------------------------------------
4928 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4929 changes of various kinds. As a consequence, a big documentation update was
4930 needed. This delayed the release for rather longer than seemed good, especially
4931 in the light of a couple of (minor) security issues. Therefore, the changes
4932 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4933 release. So 4.44 and 4.50 are in effect two different branches that both start
4936 I have left the 4.50 change log unchanged; it contains all the changes since
4937 4.43. The change log for 4.44 is below; many of its items are identical to
4938 those for 4.50. This seems to be the most sensible way to preserve the
4939 historical information.
4945 1. Minor wording change to the doc/README.SIEVE file.
4947 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4948 computation of the current number of files was incorrect.
4950 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4951 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4954 4. Give more explanation in the error message when the command for a transport
4955 filter fails to execute.
4957 5. There are several places where Exim runs a non-Exim command in a
4958 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4959 was being done only for the command run by the queryprogram router. It is
4960 now done for all such subprocesses. The other cases are: ${run, transport
4961 filters, and the commands run by the lmtp and pipe transports.
4963 6. Added CONFIGURE_GROUP build-time option.
4965 7. Some older OS have a limit of 256 on the maximum number of file
4966 descriptors. Exim was using setrlimit() to set 1000 as a large value
4967 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4968 systems. I've change it so that if it can't get 1000, it tries for 256.
4970 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4971 was an oversight, and furthermore, ever since the addition of extra
4972 controls (e.g. 4.43/32), the checks on when to allow different forms of
4973 "control" were broken. There should now be diagnostics for all cases when a
4974 control that does not make sense is encountered.
4976 9. Added the /retain_sender option to "control=submission".
4978 10. $recipients is now available in the predata ACL (oversight).
4980 11. Tidy the search cache before the fork to do a delivery from a message
4981 received from the command line. Otherwise the child will trigger a lookup
4982 failure and thereby defer the delivery if it tries to use (for example) a
4983 cached ldap connection that the parent has called unbind on.
4985 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4986 of $address_data from the recipient verification was clobbered by the
4987 sender verification.
4989 13. The value of address_data from a sender verification is now available in
4990 $sender_address_data in subsequent conditions in the ACL statement.
4992 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4994 15. Added a new option "connect=<time>" to callout options, to set a different
4997 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4998 was its contents. (It was OK if the option was not defined at all.)
5000 17. A "Completed" log line is now written for messages that are removed from
5001 the spool by the -Mrm option.
5003 18. New variables $sender_verify_failure and $recipient_verify_failure contain
5004 information about exactly what failed.
5006 19. Added -dd to debug only the daemon process.
5008 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
5009 handles timeouts, both on the server side and network timeouts. Renamed the
5010 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
5012 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
5013 It is now set to "smtps".
5015 22. $host_address is now set to the target address during the checking of
5016 ignore_target_hosts.
5018 23. When checking ignore_target_hosts for an ipliteral router, no host name was
5019 being passed; this would have caused $sender_host_name to have been used if
5020 matching the list had actually called for a host name (not very likely,
5021 since this list is usually IP addresses). A host name is now passed as
5024 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5025 code that specifies a non-restarting handler (typically sigaction() in
5026 modern systems) in an attempt to fix a rare and obscure crash bug.
5028 25. Narrowed the window for a race in the daemon that could cause it to ignore
5029 SIGCHLD signals. This is not a major problem, because they are used only to
5030 wake it up if nothing else does.
5032 26. A malformed maildirsize file could cause Exim to calculate negative values
5033 for the mailbox size or file count. Odd effects could occur as a result.
5034 The maildirsize information is now recalculated if the size or filecount
5037 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5038 support for a long time. Removed HAVE_SYS_VFS_H.
5040 28. Installed the latest version of exipick from John Jetmore.
5042 29. In an address list, if the pattern was not a regular expression, an empty
5043 subject address (from a bounce message) matched only if the pattern was an
5044 empty string. Non-empty patterns were not even tested. This was the wrong
5045 because it is perfectly reasonable to use an empty address as part of a
5046 database query. An empty address is now tested by patterns that are
5047 lookups. However, all the other forms of pattern expect the subject to
5048 contain a local part and a domain, and therefore, for them, an empty
5049 address still always fails if the pattern is not itself empty.
5051 30. Exim went into a mad DNS loop when attempting to do a callout where the
5052 host was specified on an smtp transport, and looking it up yielded more
5053 than one IP address.
5055 31. Re-factored the code for checking spool and log partition space into a
5056 function that finds that data and another that does the check. The former
5057 is then used to implement four new variables: $spool_space, $log_space,
5058 $spool_inodes, and $log_inodes.
5060 32. The RFC2047 encoding function was originally intended for short strings
5061 such as real names; it was not keeping to the 75-character limit for
5062 encoded words that the RFC imposes. It now respects the limit, and
5063 generates multiple encoded words if necessary. To be on the safe side, I
5064 have increased the buffer size for the ${rfc2047: expansion operator from
5067 33. It is now permitted to omit both strings after an "if" condition; if the
5068 condition is true, the result is "true". As before, when the second string
5069 is omitted, a false condition yields an empty string. This makes it less
5070 cumbersome to write custom ACL and router conditions.
5072 34. Failure to deliver a bounce message always caused it to be frozen, even if
5073 there was an errors_to setting on the router. The errors_to setting is now
5076 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5077 canonical form (fully expanded) before being placed in
5078 $sender_host_address.
5080 36. The table in the code that translates DNS record types into text (T_A to
5081 "A" for instance) was missing entries for NS and CNAME. It is just possible
5082 that this could have caused confusion if both these types were looked up
5083 for the same domain, because the text type is used as part of Exim's
5084 per-process caching. But the chance of anyone hitting this buglet seems
5087 37. The dnsdb lookup has been extended in a number of ways.
5089 (1) There is a new type, "zns", which walks up the domain tree until it
5090 finds some nameserver records. It should be used with care.
5092 (2) There is a new type, "mxh", which is like "mx" except that it returns
5093 just the host names, not the priorities.
5095 (3) It is now possible to give a list of domains (or IP addresses) to be
5096 looked up. The behaviour when one of the lookups defers can be
5097 controlled by a keyword.
5099 (4) It is now possible to specify the separator character for use when
5100 multiple records are returned.
5102 38. The dnslists ACL condition has been extended: it is now possible to supply
5103 a list of IP addresses and/or domains to be looked up in a particular DNS
5106 39. Added log_selector=+queue_time_overall.
5108 40. When running the queue in the test harness, wait just a tad after forking a
5109 delivery process, to get repeatability of debugging output.
5111 41. Include certificate and key file names in error message when GnuTLS fails
5112 to set them up, because the GnuTLS error message doesn't include the name
5113 of the failing file when there is a problem reading it.
5115 42. Allow both -bf and -bF in the same test run.
5117 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
5119 44. The "Exiscan patch" is now merged into the mainline Exim source.
5121 45. Sometimes the final signoff response after QUIT could fail to get
5122 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5123 < 0 before doing a fflush(). This bug looks as though it goes back to the
5124 introduction of TLS in release 3.20, but "sometimes" must have been rare
5125 because the tests only now provoked it.
5127 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
5128 (this can affect the format of dates).
5130 47. exim_tidydb, when checking for the continued existence of a message for
5131 which it has found a message-specific retry record, was not finding
5132 messages that were in split spool directories. Consequently, it was
5133 deleting retry records that should have stayed in existence.
5135 48. Steve fixed some bugs in eximstats.
5137 49. The SPA authentication driver was not abandoning authentication and moving
5138 on to the next authenticator when an expansion was forced to fail,
5139 contradicting the general specification for all authenticators. Instead it
5140 was generating a temporary error. It now behaves as specified.
5142 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
5143 (the order specifies the preference for clients). The order is now AES256,
5144 AES128, 3DES, ARCFOUR128.
5146 51. Small patch to Sieve code - explicitly set From: when generating an
5149 52. Exim crashed if a remote delivery caused a very long error message to be
5150 recorded - for instance if somebody sent an entire SpamAssassin report back
5151 as a large number of 550 error lines. This bug was coincidentally fixed by
5152 increasing the size of one of Exim's internal buffers (big_buffer) that
5153 happened as part of the Exiscan merge. However, to be on the safe side, I
5154 have made the code more robust (and fixed the comments that describe what
5157 53. Now that there can be additional text after "Completed" in log lines (if
5158 the queue_time_overall log selector is set), a one-byte patch to exigrep
5159 was needed to allow it to recognize "Completed" as not the last thing in
5162 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
5163 patch that reportedly fixes this has been added. I am not expert enough to
5164 create a test for it. This is what the patch creator wrote:
5166 "I found a little strange behaviour of ldap code when working with
5167 Windows 2003 AD Domain, where users was placed in more than one
5168 Organization Units. When I tried to give exim partial DN, the exit code
5169 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
5170 But simultaneously result of request was absolutely normal ldap result,
5171 so I produce this patch..."
5173 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
5174 REFERENCE, so I have modified the code to exclude the patch when that macro
5177 55. Some experimental protocols are using DNS PTR records for new purposes. The
5178 keys for these records are domain names, not reversed IP addresses. The
5179 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5180 leaves it alone. Component reversal etc. now happens only for IP addresses.
5183 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
5185 57. Double the size of the debug message buffer (to 2048) so that more of very
5186 long debug lines gets shown.
5188 58. The exicyclog utility now does better if the number of log files to keep
5189 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
5191 59. Two changes related to the smtp_active_hostname option:
5193 (1) $smtp_active_hostname is now available as a variable.
5194 (2) The default for smtp_banner uses $smtp_active_hostname instead
5195 of $primary_hostname.
5197 60. The host_aton() function is supposed to be passed a string that is known
5198 to be a valid IP address. However, in the case of IPv6 addresses, it was
5199 not checking this. This is a hostage to fortune. Exim now panics and dies
5200 if the condition is not met. A case was found where this could be provoked
5201 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5202 components; fortuitously, this particular loophole had already been fixed
5203 by change 4.50/55 above.
5205 If there are any other similar loopholes, the new check in host_aton()
5206 itself should stop them being exploited. The report I received stated that
5207 data on the command line could provoke the exploit when Exim was running as
5208 exim, but did not say which command line option was involved. All I could
5209 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5210 running as the user.
5213 61. There was a buffer overflow vulnerability in the SPA authentication code
5214 (which came originally from the Samba project). I have added a test to the
5215 spa_base64_to_bits() function which I hope fixes it.
5218 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
5219 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
5221 63. The daemon start-up calls getloadavg() while still root for those OS that
5222 need the first call to be done as root, but it missed one case: when
5223 deliver_queue_load_max is set with deliver_drop_privilege. This is
5224 necessary for the benefit of the queue runner, because there is no re-exec
5225 when deliver_drop_privilege is set.
5227 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
5228 This has been fixed.
5230 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5231 was in use, was not putting the data itself into the right store pool;
5232 consequently, it could be overwritten for a subsequent message in the same
5233 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5236 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
5238 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
5239 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
5240 the function were treating the return as a boolean value, which happened to
5241 work because 0=false and not-0=true, but is not correct code.
5243 68. The host_aton() function was not handling scoped IPv6 addresses (those
5244 with, for example, "%eth0" on the end) correctly.
5246 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
5247 negated items (that is, ~something) in unsigned ints. Some compilers
5248 apparently mutter when there is no cast.
5250 70. If an address verification called from an ACL failed, and did not produce a
5251 user-specific message (i.e. there was only a "system" message), nothing was
5252 put in $acl_verify_message. In this situation, it now puts the system
5255 71. Change 4.23/11 added synchronization checking at the start of an SMTP
5256 session; change 4.31/43 added the unwanted input to the log line - except
5257 that it did not do this in the start of session case. It now does.
5259 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
5260 This is wrong and can cause the other end to generate a synchronization
5261 error if it is another Exim or anything else that does the synchronization
5262 check. A QUIT command is no longer sent after a timeout.
5264 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
5265 during host lookups.
5267 74. The defer_ok option of callout verification was not working if it was used
5268 when verifying addresses in header lines, that is, for this case:
5270 verify = header_sender/callout=defer_ok
5272 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
5273 those file descriptors could be used for SMTP connections. If anything
5274 wrote to stderr (the example that came up was "warn" in embedded Perl), it
5275 could be sent to the SMTP client, causing chaos. The daemon now opens
5276 stdin, stdout, and stderr to /dev/null when it puts itself into the
5279 76. Arrange for output from Perl's "warn" command to be written to Exim's main
5280 log by default. The user can override this with suitable Perl magic.
5282 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
5283 the log message when discard triggers, was not working for the DATA ACL or
5284 for the non-SMTP ACL.
5286 78. Error message wording change in sieve.c.
5288 79. If smtp_accept_max_per_host was set, the number of connections could be
5289 restricted to fewer than expected, because the daemon was trying to set up
5290 a new connection before checking whether the processes handling previous
5291 connections had finished. The check for completed processes is now done
5292 earlier. On busy systems, this bug wouldn't be noticed because something
5293 else would have woken the daemon, and it would have reaped the completed
5296 80. If a message was submitted locally by a user whose login name contained one
5297 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
5298 It caused a spool format error. I have fixed the spool reading code. A
5299 related problem was that the "from" clause in the Received: line became
5300 illegal because of the space(s). It is now covered by ${quote_local_part.
5302 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
5305 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
5307 83. Patch from Sieve maintainer to fix unterminated string problem in
5308 "vacation" handling.
5310 84. Some minor changes to the Linux configuration files to help with other
5311 OS variants using glibc.
5313 85. One more patch for Sieve to update vacation handling to latest spec.
5316 ----------------------------------------------------
5317 See the note above about the 4.44 and 4.50 releases.
5318 ----------------------------------------------------
5324 1. Change 4.43/35 introduced a bug that caused file counts to be
5325 incorrectly computed when quota_filecount was set in an appendfile
5328 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
5329 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
5332 3. Give more explanation in the error message when the command for a transport
5333 filter fails to execute.
5335 4. There are several places where Exim runs a non-Exim command in a
5336 subprocess. The SIGUSR1 signal should be disabled for these processes. This
5337 was being done only for the command run by the queryprogram router. It is
5338 now done for all such subprocesses. The other cases are: ${run, transport
5339 filters, and the commands run by the lmtp and pipe transports.
5341 5. Some older OS have a limit of 256 on the maximum number of file
5342 descriptors. Exim was using setrlimit() to set 1000 as a large value
5343 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
5344 systems. I've change it so that if it can't get 1000, it tries for 256.
5346 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
5347 was an oversight, and furthermore, ever since the addition of extra
5348 controls (e.g. 4.43/32), the checks on when to allow different forms of
5349 "control" were broken. There should now be diagnostics for all cases when a
5350 control that does not make sense is encountered.
5352 7. $recipients is now available in the predata ACL (oversight).
5354 8. Tidy the search cache before the fork to do a delivery from a message
5355 received from the command line. Otherwise the child will trigger a lookup
5356 failure and thereby defer the delivery if it tries to use (for example) a
5357 cached ldap connection that the parent has called unbind on.
5359 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
5360 of $address_data from the recipient verification was clobbered by the
5361 sender verification.
5363 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
5364 was its contents. (It was OK if the option was not defined at all.)
5366 11. A "Completed" log line is now written for messages that are removed from
5367 the spool by the -Mrm option.
5369 12. $host_address is now set to the target address during the checking of
5370 ignore_target_hosts.
5372 13. When checking ignore_target_hosts for an ipliteral router, no host name was
5373 being passed; this would have caused $sender_host_name to have been used if
5374 matching the list had actually called for a host name (not very likely,
5375 since this list is usually IP addresses). A host name is now passed as
5378 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5379 code that specifies a non-restarting handler (typically sigaction() in
5380 modern systems) in an attempt to fix a rare and obscure crash bug.
5382 15. Narrowed the window for a race in the daemon that could cause it to ignore
5383 SIGCHLD signals. This is not a major problem, because they are used only to
5384 wake it up if nothing else does.
5386 16. A malformed maildirsize file could cause Exim to calculate negative values
5387 for the mailbox size or file count. Odd effects could occur as a result.
5388 The maildirsize information is now recalculated if the size or filecount
5391 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5392 support for a long time. Removed HAVE_SYS_VFS_H.
5394 18. Updated exipick to current release from John Jetmore.
5396 19. Allow an empty sender to be matched against a lookup in an address list.
5397 Previously the only cases considered were a regular expression, or an
5400 20. Exim went into a mad DNS lookup loop when doing a callout where the
5401 host was specified on the transport, if the DNS lookup yielded more than
5404 21. The RFC2047 encoding function was originally intended for short strings
5405 such as real names; it was not keeping to the 75-character limit for
5406 encoded words that the RFC imposes. It now respects the limit, and
5407 generates multiple encoded words if necessary. To be on the safe side, I
5408 have increased the buffer size for the ${rfc2047: expansion operator from
5411 22. Failure to deliver a bounce message always caused it to be frozen, even if
5412 there was an errors_to setting on the router. The errors_to setting is now
5415 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5416 canonical form (fully expanded) before being placed in
5417 $sender_host_address.
5419 24. Updated eximstats to version 1.33
5421 25. Include certificate and key file names in error message when GnuTLS fails
5422 to set them up, because the GnuTLS error message doesn't include the name
5423 of the failing file when there is a problem reading it.
5425 26. Expand error message when OpenSSL has problems setting up cert/key files.
5428 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
5429 (this can affect the format of dates).
5431 28. exim_tidydb, when checking for the continued existence of a message for
5432 which it has found a message-specific retry record, was not finding
5433 messages that were in split spool directories. Consequently, it was
5434 deleting retry records that should have stayed in existence.
5436 29. eximstats updated to version 1.35
5437 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
5438 1.35 - bugfix such that pie charts by volume are generated correctly
5440 30. The SPA authentication driver was not abandoning authentication and moving
5441 on to the next authenticator when an expansion was forced to fail,
5442 contradicting the general specification for all authenticators. Instead it
5443 was generating a temporary error. It now behaves as specified.
5445 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
5446 (the order specifies the preference for clients). The order is now AES256,
5447 AES128, 3DES, ARCFOUR128.
5449 31. Small patch to Sieve code - explicitly set From: when generating an
5452 32. Exim crashed if a remote delivery caused a very long error message to be
5453 recorded - for instance if somebody sent an entire SpamAssassin report back
5454 as a large number of 550 error lines. This bug was coincidentally fixed by
5455 increasing the size of one of Exim's internal buffers (big_buffer) that
5456 happened as part of the Exiscan merge. However, to be on the safe side, I
5457 have made the code more robust (and fixed the comments that describe what
5460 33. Some experimental protocols are using DNS PTR records for new purposes. The
5461 keys for these records are domain names, not reversed IP addresses. The
5462 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5463 leaves it alone. Component reversal etc. now happens only for IP addresses.
5466 34. The host_aton() function is supposed to be passed a string that is known
5467 to be a valid IP address. However, in the case of IPv6 addresses, it was
5468 not checking this. This is a hostage to fortune. Exim now panics and dies
5469 if the condition is not met. A case was found where this could be provoked
5470 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5471 components; fortuitously, this particular loophole had already been fixed
5472 by change 4.50/55 or 4.44/33 above.
5474 If there are any other similar loopholes, the new check in host_aton()
5475 itself should stop them being exploited. The report I received stated that
5476 data on the command line could provoke the exploit when Exim was running as
5477 exim, but did not say which command line option was involved. All I could
5478 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5479 running as the user.
5482 35. There was a buffer overflow vulnerability in the SPA authentication code
5483 (which came originally from the Samba project). I have added a test to the
5484 spa_base64_to_bits() function which I hope fixes it.
5487 36. The daemon start-up calls getloadavg() while still root for those OS that
5488 need the first call to be done as root, but it missed one case: when
5489 deliver_queue_load_max is set with deliver_drop_privilege. This is
5490 necessary for the benefit of the queue runner, because there is no re-exec
5491 when deliver_drop_privilege is set.
5493 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5494 was in use, was not putting the data itself into the right store pool;
5495 consequently, it could be overwritten for a subsequent message in the same
5496 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5499 38. Sometimes the final signoff response after QUIT could fail to get
5500 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5501 < 0 before doing a fflush(). This bug looks as though it goes back to the
5502 introduction of TLS in release 3.20, but "sometimes" must have been rare
5503 because the tests only now provoked it.
5509 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5510 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5511 Now it returns an integer. A number of calls were still expecting a T/F
5512 return. Fortuitously, in all cases, the tests worked in OK situations,
5513 which is the norm. However, things would have gone wrong on any write
5514 failures on the smtp file descriptor. This function is used when sending
5515 messages over SMTP and also when doing verify callouts.
5517 2. When Exim is called to do synchronous delivery of a locally submitted
5518 message (the -odf or -odi options), it no longer closes stderr before doing
5521 3. Implemented the mua_wrapper option.
5523 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5525 5. Implemented the functions header_remove(), header_testname(),
5526 header_add_at_position(), and receive_remove_recipient(), and exported them
5529 6. If an ACL "warn" statement specified the addition of headers, Exim already
5530 inserted X-ACL-Warn: at the start if there was no header name. However, it
5531 was not making this test for the second and subsequent header lines if
5532 there were newlines in the string. This meant that an invalid header could
5533 be inserted if Exim was badly configured.
5535 7. Allow an ACL "warn" statement to add header lines at the start or after all
5536 the Received: headers, as well as at the end.
5538 8. Added the rcpt_4xx retry error code.
5540 9. Added postmaster_mailfrom=xxx to callout verification option.
5542 10. Added mailfrom=xxxx to the callout verification option, for verify=
5545 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5546 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5548 12. Inserted some casts to stop certain compilers warning when using pointer
5549 differences as field lengths or precisions in printf-type calls (mostly
5550 affecting debugging statements).
5552 13. Added optional readline() support for -be (dynamically loaded).
5554 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5555 same clock tick as a message's arrival, so that its received time was the
5556 same as the "first fail" time on the retry record, and that message
5557 remained on the queue past the ultimate address timeout, every queue runner
5558 would try a delivery (because it was past the ultimate address timeout) but
5559 after another failure, the ultimate address timeout, which should have then
5560 bounced the address, did not kick in. This was a "< instead of <=" error;
5561 in most cases the first failure would have been in the next clock tick
5562 after the received time, and all would be well.
5564 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5565 being recognized when the domain list was tested by the match_domain
5566 condition in an expansion string.
5568 16. Added the ${str2b64: operator.
5570 17. Exim was always calling setrlimit() to set a large limit for the number of
5571 processes, without checking whether the existing limit was already
5572 adequate. (It did check for the limit on file descriptors.) Furthermore,
5573 errors from getrlimit() and setrlimit() were being ignored. Now they are
5574 logged to the main and panic logs, but Exim does carry on, to try to do its
5575 job under whatever limits there are.
5577 18. Imported PCRE 5.0.
5579 19. Trivial typo in log message " temporarily refused connection" (the leading
5582 20. If the log selector return_path_on_delivery was set and an address was
5583 redirected to /dev/null, the delivery process crashed because it assumed
5584 that a return path would always be set for a "successful" delivery. In this
5585 case, the whole delivery is bypassed as an optimization, and therefore no
5588 21. Internal re-arrangement: the function for sending a challenge and reading
5589 a response while authentication was assuming a zero-terminated challenge
5590 string. It's now changed to take a pointer and a length, to allow for
5591 binary data in such strings.
5593 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5595 23. Exim was not respecting finduser_retries when seeking the login of the
5596 uid under which it was called; it was always trying 10 times. (The default
5597 setting of finduser_retries is zero.) Also, it was sleeping after the final
5598 failure, which is pointless.
5600 24. Implemented tls_on_connect_ports.
5602 25. Implemented acl_smtp_predata.
5604 26. If the domain in control=submission is set empty, Exim assumes that the
5605 authenticated id is a complete email address when it generates From: or
5606 Sender: header lines.
5608 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5609 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5610 chown and chgrp in /bin and hostname in /usr/bin.
5612 28. Exim was keeping the "process log" file open after each use, just as it
5613 does for the main log. This opens the possibility of it remaining open for
5614 long periods when the USR1 signal hits a daemon. Occasional processlog
5615 errors were reported, that could have been caused by this. Anyway, it seems
5616 much more sensible not to leave this file open at all, so that is what now
5619 29. The long-running daemon process does not normally write to the log once it
5620 has entered its main loop, and it closes the log before doing so. This is
5621 so that log files can straightforwardly be renamed and moved. However,
5622 there are a couple of unusual error situations where the daemon does write
5623 log entries, and I had neglected to close the log afterwards.
5625 30. The text of an SMTP error response that was received during a remote
5626 delivery was being truncated at 512 bytes. This is too short for some of
5627 the long messages that one sometimes sees. I've increased the limit to
5630 31. It is now possible to make retry rules that apply only when a message has a
5631 specific sender, in particular, an empty sender.
5633 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5634 it possible to be selective about when SMTP synchronization is enforced.
5636 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5638 32. Implemented hosts_connection_nolog.
5640 33. Added an ACL for QUIT.
5642 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5645 35. Added mailbox_size and mailbox_filecount to appendfile.
5647 36. Added control = no_multiline_responses to ACLs.
5649 37. There was a bug in the logic of the code that waits for the clock to tick
5650 in the case where the clock went backwards by a substantial amount such
5651 that the microsecond fraction of "now" was more than the microsecond
5652 fraction of "then" (but the whole seconds number was less).
5654 38. Added support for the libradius Radius client library this is found on
5655 FreeBSD (previously only the radiusclient library was supported).
5661 1. When certain lookups returned multiple values in the form name=value, the
5662 quoting of the values was not always being done properly. Specifically:
5663 (a) If the value started with a double quote, but contained no whitespace,
5665 (b) If the value contained whitespace other than a space character (i.e.
5666 tabs or newlines or carriage returns) it was not quoted.
5667 This fix has been applied to the mysql and pgsql lookups by writing a
5668 separate quoting function and calling it from the lookup code. The fix
5669 should probably also be applied to nisplus, ibase and oracle lookups, but
5670 since I cannot test any of those, I have not disturbed their existing code.
5672 2. A hit in the callout cache for a specific address caused a log line with no
5673 reason for rejecting RCPT. Now it says "Previous (cached) callout
5674 verification failure".
5676 3. There was an off-by-one bug in the queryprogram router. An over-long
5677 return line was truncated at 256 instead of 255 characters, thereby
5678 overflowing its buffer with the terminating zero. As well as fixing this, I
5679 have increased the buffer size to 1024 (and made a note to document this).
5681 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5682 when Exim is waiting for an SMTP response from a remote server, Exim
5683 restarts its select() call on the socket, thereby resetting its timeout.
5684 This is not a problem when such interrupts are rare. Somebody set up a cron
5685 job to run exiwhat every 2 minutes, which is less than the normal select()
5686 timeout (5 or 10 minutes). This meant that the select() timeout never
5687 kicked in because it was always reset. I have fixed this by comparing the
5688 time when an interrupt arrives with the time at the start of the first call
5689 to select(). If more time than the timeout has elapsed, the interrupt is
5690 treated as a timeout.
5692 5. Some internal re-factoring in preparation for the addition of Sieve
5693 extensions (by MH). In particular, the "personal" test is moved to a
5694 separate function, and given an option for scanning Cc: and Bcc: (which is
5695 not set for Exim filters).
5697 6. When Exim created an email address using the login of the caller as the
5698 local part (e.g. when creating a From: or Sender: header line), it was not
5699 quoting the local part when it contained special characters such as @.
5701 7. Installed new OpenBSD configuration files.
5703 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5704 try to make them clearer.
5706 9. Callout options, other than the timeout value, were being ignored when
5707 verifying sender addresses in header lines. For example, when using
5709 verify = header_sender/callout=no_cache
5711 the cache was (incorrectly) being used.
5713 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5714 only the Cygwin environment.
5716 11. When return_path_on_delivery was set as a log selector, if different remote
5717 addresses in the same message used different return paths and parallel
5718 remote delivery occurred, the wrong values would sometimes be logged.
5719 (Whenever a remote delivery process finished, the return path value from
5720 the most recently started remote delivery process was logged.)
5722 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5723 lines when AUTH and/or TLS are in use. This is the "received protocol"
5724 field. Exim used to use "asmtp" for authenticated SMTP, without any
5725 indication (in the protocol name) for TLS use. Now it follows the RFC and
5726 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5727 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5728 names appear in log lines as well as in Received: header lines.
5730 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5731 extensions, and comparison tests, and to fix some bugs.
5733 14. Changes to the "personal" filter test:
5735 (1) The test was buggy in that it was just doing the equivalent of
5736 "contains" tests on header lines. For example, if a user's address was
5737 anne@some.where, the "personal" test would incorrectly be true for
5739 To: susanne@some.where
5741 This test is now done by extracting each address from the header in turn,
5742 and checking the entire address. Other tests that are part of "personal"
5743 are now done using regular expressions (for example, to check local parts
5744 of addresses in From: header lines).
5746 (2) The list of non-personal local parts in From: addresses has been
5747 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5748 taken from the Sieve specification recommendations.
5750 (3) If the message contains any header line starting with "List-" it is
5751 treated as non-personal.
5753 (4) The test for "circular" in the Subject: header line has been removed
5754 because it now seems ill-conceived.
5756 15. Minor typos in src/EDITME comments corrected.
5758 16. Installed latest exipick from John Jetmore.
5760 17. If headers_add on a router specified a text string that was too long for
5761 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5762 of string_sprintf() is now avoided.
5764 18. $message_body_size was not set (it was always zero) when running the DATA
5765 ACL and the local_scan() function.
5767 19. For the "mail" command in an Exim filter, no default was being set for
5768 the once_repeat time, causing a random time value to be used if "once" was
5769 specified. (If the value happened to be <= 0, no repeat happened.) The
5770 default is now 0s, meaning "never repeat". The "vacation" command was OK
5771 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5772 (I found it when inspecting the code).
5774 20. There is now an overall timeout for performing a callout verification. It
5775 defaults to 4 times the callout timeout, which applies to individual SMTP
5776 commands during the callout. The overall timeout applies when there is more
5777 than one host that can be tried. The timeout is checked before trying the
5778 next host. This prevents very long delays if there are a large number of
5779 hosts and all are timing out (e.g. when the network connections are timing
5780 out). The value of the overall timeout can be changed by specifying an
5781 additional sub-option for "callout", called "maxwait". For example:
5783 verify = sender/callout=5s,maxwait=20s
5785 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5786 to the end before writing, but this should make it even safer).
5788 22. Exim was forgetting that it had advertised PIPELINING for the second and
5789 subsequent messages on an SMTP connection. It was also not resetting its
5790 memory on STARTTLS and an internal HELO.
5792 23. When Exim logs an SMTP synchronization error within a session, it now
5793 records whether PIPELINING has been advertised or not.
5795 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5796 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5797 rather than long int.
5799 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5801 26. Added the never_mail option to autoreply.
5807 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5808 crash if the getsockname() call failed; this can happen if a connection is
5809 closed very soon after it is established. The problem was simply in the
5810 order in which certain operations were done, causing Exim to try to write
5811 to the SMTP stream before it had set up the file descriptor. The bug has
5812 been fixed by making things happen in the correct order.
5818 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5819 before the connection was closed, thus losing the rejection response.
5821 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5822 some early Solaris releases, but causes trouble in current releases where
5823 socklen_t is defined.
5825 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5828 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5831 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5832 that releasing the top of it at the end releases what was used for sub-
5833 expansions (unless the block got too big). However, discard this block if
5834 the first thing is a variable or header, so that we can use its block when
5835 it is dynamic (useful for very large $message_headers, for example).
5837 6. Lookups now cache *every* query, not just the most recent. A new, separate
5838 store pool is used for this. It can be recovered when all lookup caches are
5839 flushed. Lookups now release memory at the end of their result strings.
5840 This has involved some general refactoring of the lookup sources.
5842 7. Some code has been added to the store_xxx() functions to reduce the amount
5843 of flapping under certain conditions.
5845 8. log_incoming_interface used to affect only the <= reception log lines. Now
5846 it causes the local interface and port to be added to several more SMTP log
5847 lines, for example "SMTP connection from", and rejection lines.
5849 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5851 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5853 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5854 could be overwritten at the end of the current message (or the start of a
5855 new message if it was set in a HELO ACL). The value is now preserved for
5856 the duration of the SMTP connection.
5858 12. If a transport had a headers_rewrite setting, and a matching header line
5859 contained an unqualified address, that address was qualified, even if it
5860 did not match any rewriting rules. The underlying bug was that the values
5861 of the flags that permit the existence of unqualified sender and recipient
5862 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5863 non-local messages, and by -bnq for local messages) were not being
5864 preserved with the message after it was received.
5866 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5867 "next input=" as part of the text comprising the host identity instead of
5868 the correct text. The code was using the same buffer for two different
5869 strings. However, depending on which order the printing function evaluated
5870 its arguments, the bug did not always show up. Under Linux, for example, my
5871 test suite worked just fine.
5873 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5874 doesn't work with some older versions of Perl. It has been changed to "my",
5875 which in any case is probably the better facility to use.
5877 15. A really picky compiler found some instances of statements for creating
5878 error messages that either had too many or two few arguments for the format
5881 16. The size of the buffer for calls to the DNS resolver has been increased
5882 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5883 for addresses that have a lot of PTR records. This alleviates a problem; it
5884 does not fully solve it.
5886 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5887 buffer now truncates the list and logs the incident, which is the same
5888 action as happens when Exim is looking up a host name and its aliases.
5889 Previously in this situation something unpredictable would happen;
5890 sometimes it was "internal error: store_reset failed".
5892 18. If a server dropped the connection unexpectedly when an Exim client was
5893 using GnuTLS and trying to read a response, the client delivery process
5894 crashed while trying to generate an error log message.
5896 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5897 string, for example:
5899 warn message = H1: something\nH2: something
5901 the text was added as a single header line from Exim's point of view
5902 though it ended up OK in the delivered message. However, searching for the
5903 second and subsequent header lines using $h_h2: did not work. This has been
5904 fixed. Similarly, if a system filter added multiple headers in this way,
5905 the routers could not see them.
5907 20. Expanded the error message when iplsearch is called with an invalid key to
5908 suggest using net-iplsearch in a host list.
5910 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5911 ACLs are no longer actually imposed (and a message to that effect is
5914 22. If a "gecos" field in a passwd entry contained escaped characters, in
5915 particular, if it contained a \" sequence, Exim got it wrong when building
5916 a From: or a Sender: header from that name. A second bug also caused
5917 incorrect handling when an unquoted " was present following a character
5918 that needed quoting.
5920 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5921 was not being matched caselessly.
5923 24. Arranged for all hyphens in the exim.8 source to be escaped with
5926 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5927 properly. Recipient callout cache records were still being keyed to include
5928 the sender, even when use_sender was set false. This led to far more
5929 callouts that were necessary. The sender is no longer included in the key
5930 when use_sender is false.
5932 26. Added "control = submission" modifier to ACLs.
5934 27. Added the ${base62d: operator to decode base 62 numbers.
5936 28. dnsdb lookups can now access SRV records.
5938 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5939 the configuration file.
5941 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5942 (-v) mode. This makes the output for a verbose queue run more intelligible.
5944 31. Added a use_postmaster feature to recipient callouts.
5946 32. Added the $body_zerocount variable, containing the number of binary zero
5947 bytes in the message body.
5949 33. The time of last modification of the "new" subdirectory is now used as the
5950 "mailbox time last read" when there is a quota error for a maildir
5953 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5955 35. Added +ignore_unknown as a special item in host lists.
5957 36. Code for decoding IPv6 addresses in host lists is now included, even if
5958 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5959 address was recognized as an IP address, but was then not correctly decoded
5960 into binary, causing unexpected and incorrect effects when compared with
5967 1. Very minor rewording of debugging text in manualroute to say "list of
5968 hosts" instead of "hostlist".
5970 2. If verify=header_syntax was set, and a header line with an unqualified
5971 address (no domain) and a large number of spaces between the end of the
5972 name and the colon was received, the reception process suffered a buffer
5973 overflow, and (when I tested it) crashed. This was caused by some obsolete
5974 code that should have been removed. The fix is to remove it!
5976 3. When running in the test harness, delay a bit after writing a bounce
5977 message to get a bit more predictability in the log output.
5979 4. Added a call to search_tidyup() just before forking a reception process. In
5980 theory, someone could use a lookup in the expansion of smtp_accept_max_
5981 per_host which, without the tidyup, could leave open a database connection.
5983 5. Added the variables $recipient_data and $sender_data which get set from a
5984 lookup success in an ACL "recipients" or "senders" condition, or a router
5985 "senders" option, similar to $domain_data and $local_part_data.
5987 6. Moved the writing of debug_print from before to after the "senders" test
5990 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5991 problems for message scanning, either using a data ACL, or using
5992 local_scan() because the Received: header was not generated till after they
5993 were called (in order to set the time as the time of reception completion).
5994 I have revised the way this works. The header is now generated after the
5995 body is received, but before the ACL or local_scan() are called. After they
5996 are run, the timestamp in the header is updated.
6002 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
6003 before starting a queue runner without re-exec. This happened only when
6004 deliver_drop_privilege was set or when the Exim user was set to root. The
6005 effect of the bug was that timeouts during subsequent deliveries caused
6006 crashes instead of being properly handled. The handler is now left at its
6007 default (and expected) setting.
6009 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
6010 message, again when deliver_drop_privilege is set or Exim is run as root.
6011 The bug described in (1) was not present in this case, but the tidying up
6012 of the other signals was missing. I have made the two cases consistent.
6014 3. The ignore_target_hosts setting on a manualroute router was being ignored
6015 for hosts that were looked up using the /MX notation.
6017 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
6020 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
6021 operated on the sender address. After changing the $sender_address to <>
6022 for the sender address verify, Exim was re-instated it as the original
6023 (before rewriting) address, but remembering that it had rewritten it, so it
6024 wasn't rewriting it again. This bug also had the effect of breaking the
6025 sender address verification caching when the sender address was rewritten.
6027 6. The ignore_target_hosts option was being ignored by the ipliteral router.
6028 This has been changed so that if the ip literal address matches
6029 ignore_target_hosts, the router declines.
6031 7. Added expansion conditions match_domain, match_address, and match_local_
6032 part (NOT match_host).
6034 8. The placeholder for the Received: header didn't have a length field set.
6036 9. Added code to Exim itself and to exim_lock to test for a specific race
6037 condition that could lead to file corruption when using MBX delivery. The
6038 issue is with the lockfile that is created in /tmp. If this file is removed
6039 after a process has opened it but before that process has acquired a lock,
6040 there is the potential for a second process to recreate the file and also
6041 acquire a lock. This could lead to two Exim processes writing to the file
6042 at the same time. The added code performs the same test as UW imapd; it
6043 checks after acquiring the lock that its file descriptor still refers to
6044 the same named file.
6046 10. The buffer for building added header lines was of fixed size, 8192 bytes.
6047 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
6050 11. Added the smtp_active_hostname option. If used, this will typically be made
6051 to depend on the incoming interface address. Because $interface_address is
6052 not set up until the daemon has forked a reception process, error responses
6053 that can happen earlier (such as "too many connections") no longer contain
6056 12. If an expansion in a condition on a "warn" statement fails because a lookup
6057 defers, the "warn" statement is abandoned, and the next ACL statement is
6058 processed. Previously this caused the whole ACL to be aborted.
6060 13. Added the iplsearch lookup type.
6062 14. Added ident_timeout as a log selector.
6064 15. Added tls_certificate_verified as a log selector.
6066 16. Added a global option tls_require_ciphers (compare the smtp transport
6067 option of the same name). This controls incoming TLS connections.
6069 17. I finally figured out how to make tls_require_ciphers do a similar thing
6070 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
6071 before starting the TLS session.
6073 18. Tabs are now shown as \t in -bP output.
6075 19. If the log selector return_path_on_delivery was set, Exim crashed when
6076 bouncing a message because it had too many Received: header lines.
6078 20. If two routers both had headers_remove settings, and the first one included
6079 a superfluous trailing colon, the final name in the first list and the
6080 first name in the second list were incorrectly joined into one item (with a
6081 colon in the middle).
6087 1. Added -C and -D options to the exinext utility, mainly to make it easier
6088 to include in the automated testing, but these could be helpful when
6089 multiple configurations are in use.
6091 2. The exinext utility was not formatting the output nicely when there was
6092 an alternate port involved in the retry record key, nor when there was a
6093 message id as well (for retries that were specific to a specific message
6094 and a specific host). It was also confused by IPv6 addresses, because of
6095 the additional colons they contain. I have fixed the IPv4 problem, and
6096 patched it up to do a reasonable job for IPv6.
6098 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
6099 delivery, the log line now contains "pipelined" if PIPELINING was used.
6101 4. An SMTP transport process used to panic and die if the bind() call to set
6102 an explicit outgoing interface failed. This has been changed; it is now
6103 treated in the same way as a connect() failure.
6105 5. A reference to $sender_host_name in the part of a conditional expansion
6106 that was being skipped was still causing a DNS lookup. This no longer
6109 6. The def: expansion condition was not recognizing references to header lines
6110 that used bh_ and bheader_.
6112 7. Added the _cache feature to named lists.
6114 8. The code for checking quota_filecount in the appendfile transport was
6115 allowing one more file than it should have been.
6117 9. For compatibility with Sendmail, the command line option
6125 and sets the incoming protocol and host name (for trusted callers). The
6126 host name and its colon can be omitted when only the protocol is to be set.
6127 Note the Exim already has two private options, -pd and -ps, that refer to
6128 embedded Perl. It is therefore impossible to set a protocol value of "d" or
6129 "s", but I don't think that's a major issue.
6131 10. A number of refactoring changes to the code, none of which should affect
6134 (a) The number of logging options was getting close to filling up the
6135 32-bit word that was used as a bit map. I have split them into two classes:
6136 those that are passed in the argument to log_write(), and those that are
6137 only ever tested independently outside of that function. These are now in
6138 separate 32-bit words, so there is plenty of room for expansion again.
6139 There is no change in the user interface or the logging behaviour.
6141 (b) When building, for example, log lines, the code previously used a
6142 macro that called string_cat() twice, in order to add two strings. This is
6143 not really sufficiently general. Furthermore, there was one instance where
6144 it was actually wrong because one of the argument was used twice, and in
6145 one call a function was used. (As it happened, calling the function twice
6146 did not affect the overall behaviour.) The macro has been replaced by a
6147 function that can join an arbitrary number of extra strings onto a growing
6150 (c) The code for expansion conditions now uses a table and a binary chop
6151 instead of a serial search (which was left over from when there were very
6152 few conditions). Also, it now recognizes conditions like "pam" even when
6153 the relevant support is not compiled in: a suitably worded error message is
6154 given if an attempt is made to use such a condition.
6156 11. Added ${time_interval:xxxxx}.
6158 12. A bug was causing one of the ddress fields not to be passed back correctly
6159 from remote delivery subprocesses. The field in question was not being
6160 subsequently used, so this caused to problems in practice.
6162 13. Added new log selectors queue_time and deliver_time.
6164 14. Might have fixed a bug in maildirsizefile handling that threw up
6165 "unexpected character" debug warnings, and recalculated the data
6166 unnecessarily. In any case, I expanded the warning message to give more
6169 15. Added the message "Restricted characters in address" to the statements in
6170 the default ACL that block characters like @ and % in local parts.
6172 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
6173 Three changes have been made:
6175 (a) There was a serious bug; a negative response to MAIL caused the whole
6176 recipient domain to be cached as invalid, thereby blocking all messages
6177 to all local parts at the same domain, from all senders. This bug has
6178 been fixed. The domain is no longer cached after a negative response to
6179 MAIL if the sender used is not empty.
6181 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
6184 (c) A new callout option, "use_sender" has been added for people who want
6185 the modified behaviour.
6191 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
6194 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
6195 indeed breaks things for older releases.
6197 3. Added additional logging to the case where there is a problem reading data
6198 from a filter that is running in a subprocess using a pipe, in order to
6199 try to track down a specific problem.
6201 4. Testing facility fudge: when running in the test harness and attempting
6202 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
6203 getting "No route to host". Convert this to a timeout.
6205 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
6208 6. Some OS don't have socklen_t but use size_t instead. This affects the
6209 fifth argument of getsockopt() amongst other things. This is now
6210 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
6211 can be set for individual OS. I have set it for SunOS5, OSF1, and
6212 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
6213 some earlier ones do not.
6215 7. Change 4.30/15 was not doing the test caselessly.
6217 8. The standard form for an IPv6 address literal was being rejected by address
6218 parsing in, for example, MAIL and RCPT commands. An example of this kind of
6219 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6220 this, as well as the form without the "IPv6" on the front (but only when
6221 address literals are enabled, of course).
6223 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
6225 10. Exim crashed if a message with an empty sender address specified by -f
6226 encountered a router with an errors_to setting. This could be provoked only
6227 by a command such as
6231 where an empty string was supplied; "<>" did not hit this bug.
6233 11. Installed PCRE release 4.5.
6235 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
6236 remained set. It is now erased.
6238 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
6239 times from message ids (which are base 36 rather than the normal 62).
6241 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
6242 were being counted as actual protocol errors, and logged if the log
6243 selector +smtp_protocol_error was set. One cannot be perfect in this test,
6244 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
6245 and DATA following a set of rejected RCPTs do not count as protocol errors.
6246 In other words, Exim assumes they were pipelined, though this may not
6247 actually be the case. Of course, in all cases the client gets an
6248 appropriate error code.
6250 15. If a lookup fails in an ACL condition, a message about the failure may
6251 be available; it is used if testing the ACL cannot continue, because most
6252 such messages specify what the cause of the deferral is. However, some
6253 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
6254 that caused an old message to be retained and used if a later statement
6255 caused a defer, replacing the real cause of the deferral.
6257 16. If an IP address had so many PTR records that the DNS lookup buffer
6258 was not large enough to hold them, Exim could crash while trying to process
6259 the truncated data. It now detects and logs this case.
6261 17. Further to 4.21/58, another change has been made: if (and only if) the
6262 first line of a message (the first header line) ends with CRLF, a bare LF
6263 in a subsequent header line has a space inserted after it, so as not to
6264 terminate the header.
6266 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
6267 unnecessarily, used atoi() instead of strtol(), and didn't check the
6268 termination when getting file sizes from file names by regex.
6270 19. Completely re-implemented the support for maildirsize files, in the light
6271 of a number of problems with the previous contributed implementation
6272 (4.30/29). In particular:
6274 . If the quota is zero, the maildirsize file is maintained, but no quota is
6277 . If the maildir directory does not exist, it is created before any attempt
6278 to write a maildirsize file.
6280 . The quota value in the file is just a cache; if the quota is changed in
6281 the transport, the new value overrides.
6283 . A regular expression is available for excluding directories from the
6286 20. The autoreply transport checks the characters in options that define the
6287 message's headers; it allows continued headers, but it was checking with
6288 isspace() after an embedded newline instead of explicitly looking for a
6291 21. If all the "regular" hosts to which an address was routed had passed their
6292 expiry times, and had not reached their retry times, the address was
6293 bounced, even if fallback hosts were defined. Now Exim should go on to try
6296 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
6297 equivalent code in the SMTP transport. Some hosts send humungous responses
6298 to HELO/EHLO, more than 1024 it seems.
6300 23. Refactoring: code in filter.c used (void *) for "any old type" but this
6301 gives compiler warnings in some environments. I've now done it "properly",
6304 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
6305 (because of problems with the built-in one) was declared to return uschar *
6306 instead of char *, causing compiler failure.
6308 25. Fixed a file descriptor leak when processing alias/forward files.
6310 26. Fixed a minor format string issue in dbfn.c.
6312 27. Typo in exim.c: ("dmbnz" for "dbmnz").
6314 28. If a filter file refered to $h_xxx or $message_headers, and the headers
6315 contained RFC 2047 "words", Exim's memory could, under certain conditions,
6318 29. When a sender address is verified, it is cached, to save repeating the test
6319 when there is more than one recipient in a message. However, when the
6320 verification involves a callout, it is possible for different callout
6321 options to be set for different recipients. It is too complicated to keep
6322 track of this in the cache, so now Exim always runs a verification when a
6323 callout is required, relying on the callout cache for the optimization.
6324 The overhead is duplication of the address routing, but this should not be
6327 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
6328 to be verified with callout=postmaster, and the main callout worked but the
6329 postmaster check failed, the verification correctly failed. However, if a
6330 subsequent RCPT command asked for sender verification *without* the
6331 postmaster check, incorrect caching caused this verification also to fail,
6334 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
6335 it was not caching the DNS options (qualify_single, search_parents) that
6336 were used when the lookup failed. A subsequent lookup with different
6337 options therefore always gave the same answer, though there were cases
6338 where it should not have. (Example: a "domains = !$mx_any" option on a
6339 dnslookup router: the "domains" option is always processed without any
6340 widening, but the router might have qualify_single set.) Now Exim uses the
6341 cached value only when the same options are set.
6343 32. Added John Jetmore's "exipick" utility to the distribution.
6345 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
6346 than a timeout (e.g. a certificate is required, and is not provided), an
6347 Exim server now closes the connection immediately. Previously it waited for
6348 the client to close - but if the client is SSL, it seems that they each
6349 wait for each other, leading to a delay before one of them times out.
6351 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
6352 maintained 0.8.x compatibility because I don't think many are using it, and
6353 it is clearly obsolete.
6355 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
6358 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
6359 client certificate was expired. A simple patch fixes this, though I don't
6360 understand the full logic of why the verify callback is called multiple
6363 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
6364 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
6365 which causes problems with some clients (such as the Certicom SSL Plus
6366 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
6367 disables the coutermeasure allowing Eudora to connect."
6369 38. Exim was not checking that a write() to a log file succeeded. This could
6370 lead to Bad Things if a log got too big, in particular if it hit a file
6371 size limit. Exim now panics and dies if it cannot write to a log file, just
6372 as it does if it cannot open a log file.
6374 39. Modified OS/Makefile-Linux so that it now contains
6376 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
6378 The two -D definitions ensure that Exim is compiled with large file
6379 support, which makes it possible to handle log files that are bigger than
6382 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
6383 instance) a domain was checked against a named list that involved a lookup,
6384 causing $domain_data to be set, then another domain was checked against the
6385 same list, then the first domain was re-checked, the value of $domain_data
6386 after the final check could be wrong. In particular, if the second check
6387 failed, it could be set empty. This bug probably also applied to
6390 41. The strip_trailing_dot option was not being applied to the address given
6391 with the -f command-line option.
6393 42. The code for reading a message's header from the spool was incrementing
6394 $received_count, but never initializing it. This meant that the value was
6395 incorrect (doubled) while delivering a message in the same process in which
6396 it was received. In the most common configuration of Exim, this never
6397 happens - a fresh exec is done - but it can happen when
6398 deliver_drop_privilege is set.
6400 43. When Exim logs an SMTP synchronization error - client data sent too soon -
6401 it now includes up to 150 characters of the unexpected data in the log
6404 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
6405 and building data strings. The size of both of these buffers was 10 000
6406 bytes - far larger than anybody would *ever* want, thought I. Needless to
6407 say, somebody hit the limit. I have increased the maximum line length to
6408 20 000 and the maximum data length of concatenated lines to 100 000. I have
6409 also fixed two bugs, because there was no checking on these buffers. Tsk,
6410 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
6411 buffer is too small.
6413 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
6414 lsearch lookups. Now it does.
6416 46. When parsing a route_list item in a manualroute router, a fixed-length
6417 buffer was used for the list of hosts. I made this 1024 bytes long,
6418 thinking that nobody would ever have a list of hosts that long. Wrong.
6419 Somebody had a whole pile of complicated expansion conditions, and the
6420 string was silently truncated, leading to an expansion error. It turns out
6421 that it is easier to change to an unlimited length (owing to other changes
6422 that have happened since this code was originally written) than to build
6423 structure for giving a limitation error. The length of the item that
6424 expands into the list of hosts is now unlimited.
6426 47. The lsearch lookup could not handle data where the length of text line was
6427 more than 4095 characters. Such lines were truncated, leading to shortened
6428 data being returned. It should now handle lines of any length.
6430 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
6431 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
6434 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
6435 replacing strings such as BIN_DIRECTORY in a source file: the replacement
6436 no longer happens in comment lines. A list of replacements is now placed
6437 at the head of all of the source files, except those whose only change is
6438 to replace PERL_COMMAND in the very first #! line.
6440 50. Replaced the slow insertion sort in queue.c, for sorting the list of
6441 messages on the queue, with a bottom-up merge sort, using code contributed
6442 by Michael Haardt. This should make operations like -bp somewhat faster on
6443 large queues. It won't affect queue runners, except when queue_run_in_order
6446 51. Installed eximstats 1.31 in the distribution.
6448 52. Added support for SRV lookups to the dnslookup router.
6450 53. If an ACL referred to $message_body or $message_body_end, the value was not
6451 reset for any messages that followed in the same SMTP session.
6453 54. The store-handling optimization for building very long strings was not
6454 differentiating between the different store pools. I don't think this
6455 actually made any difference in practice, but I've tidied it.
6457 55. While running the routers to verify a sender address, $sender_address
6458 was still set to the sender address. This is wrong, because when routing to
6459 send a bounce to the sender, it would be empty. Therefore, I have changed
6460 it so that, while verifying a sender address, $sender_address is set to <>.
6461 (There is no change to what happens when verifying a recipient address.)
6463 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
6464 target A or AAAA records (if not already returned) without resetting the
6465 qualify_single or search_parents options of the DNS resolver. These are
6466 inappropriate in this case because the targets of MX and SRV records must
6467 be FQDNs. A broken DNS record could cause trouble if it happened to have a
6468 target that, when qualified, matched something in the local domain. These
6469 two options are now turned off when doing these lookups.
6471 57. It seems that at least some releases of Reiserfs (which does not have the
6472 concept of a fixed number of inodes) returns zero and not -1 for the
6473 number of available inodes. This interacted badly with check_spool_inodes,
6474 which assumed that -1 was the "no such thing" setting. What I have done is
6475 to check that the total number of inodes is greater than zero before doing
6476 the test of how many are available.
6478 58. When a "warn" ACL statement has a log_message modifier, the message is
6479 remembered, and not repeated. This is to avoid a lot of repetition when a
6480 message has many recipients that cause the same warning to be written.
6481 However, Exim was preserving the list of already written lines for an
6482 entire SMTP session, which doesn't seem right. The memory is now reset if a
6483 new message is started.
6485 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6486 debugging output unless log_rewrite was also set.
6488 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6489 of (int)(handle) when we know that handle contains (void *)(-1).
6491 61. The Exim daemon panic-logs an error return when it closes the incoming
6492 connection. However "connection reset by peer" seems to be common, and
6493 isn't really an error worthy of noting specially, so that particular error
6496 62. When Exim is trying to find all the local interfaces, it used to panic and
6497 die if the ioctl to get the interface flags failed. However, it seems that
6498 on at least one OS (Solaris 9) it is possible to have an interface that is
6499 included in the list of interfaces, but for which you get a failure error
6500 for this call. This happens when the interface is not "plumbed" into a
6501 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6502 failure of the "get flags" call assumes that the interface is down.
6504 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6505 makes life easier for people who are doing arithmetic on fields extracted
6506 from dates, where you often get leading zeros that should not be
6507 interpreted as octal.
6509 64. Added qualify_domain to the redirect router, to override the global
6512 65. If a pathologically long header line contained very many addresses (the
6513 report of this problem mentioned 10 000) and each of them was rewritten,
6514 Exim could use up a very large amount of memory. (It kept on making new
6515 copies of the header line as it rewrote, and never released the old ones.)
6516 At the expense of a bit more processing, the header rewriting function has
6517 been changed so that it no longer eats memory in this way.
6519 66. The generation of the Received: header has been moved from the time that a
6520 message starts to be received, to the time that it finishes. The timestamp
6521 in the Received: header should now be very close to that of the <= log
6522 line. There are two side-effects of this change:
6524 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6525 logged header lines no longer include the local Received: line, because
6526 it has not yet been created. The same applies to a copy of the message
6527 that is returned to a non-SMTP sender when a message is rejected.
6529 (b) When a filter file is tested using -bf, no additional Received: header
6530 is added to the test message. After some thought, I decided that this
6533 This change does not affect the value of $received_for. It is still set
6534 after address rewriting, but before local_scan() is called.
6536 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6538 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6539 gave an unhelpful panic error message, and a defer error. I have managed to
6540 change this behaviour so that it now rejects any supplied certificate,
6541 which seems right, as the list of acceptable certificates is empty.
6543 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6544 gave an unhelpful defer error. I have not managed to make this reject any
6545 supplied certificates, but the error message it gives is "no certificate
6546 supplied", which is not helpful.
6548 70. exigrep's output now also includes lines that are not associated with any
6549 message, but which match the given pattern. Implemented by a patch from
6550 Martin Sluka, which also tidied up the Perl a bit.
6552 71. Recipient callout verification, like sender verification, was using <> in
6553 the MAIL FROM command. This isn't really the right thing, since the actual
6554 sender may affect whether the remote host accepts the recipient or not. I
6555 have changed it to use the actual sender in the callout; this means that
6556 the cache record is now keyed on a recipient/sender pair, not just the
6557 recipient address. There doesn't seem to be a real danger of callout loops,
6558 since a callout by the remote host to check the sender would use <>.
6559 [SEE ABOVE: changed after hitting problems.]
6561 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6562 temporary errors. However, in the case of such a code being given after
6563 the end of a data transmission (i.e. after ".") Exim was failing to write
6564 a retry record for the message. (Yes, there was some broken host that was
6565 actually sending 8xx at this point.)
6567 73. An unknown lookup type in a host list could cause Exim to panic-die when
6568 the list was checked. (An example that provoked this was putting <; in the
6569 middle of a list instead of at the start.) If this happened during a DATA
6570 ACL check, a -D file could be left lying around. This kind of configuration
6571 error no longer causes Exim to die; instead it causes a defer error. The
6572 incident is still logged to the main and panic logs.
6574 74. Buglet left over from Exim 3 conversion. The message "too many messages
6575 in one connection" was written to the rejectlog but not the mainlog, except
6576 when address rewriting (yes!) was being logged.
6578 75. Added write_rejectlog option.
6580 76. When a system filter was run not as root (that is, when system_filter_user
6581 was set), the values of the $n variables were not being returned to the
6582 main process; thus, they were not subsequently available in the $sn
6585 77. Added +return_path_on_delivery log selector.
6587 78. A connection timeout was being treated differently from recipients deferred
6588 when testing hosts_max_try with a message that was older than the host's
6589 retry timeout. (The host should not be counted, thus allowing all hosts to
6590 be tried at least once before bouncing.) This may have been the cause of an
6591 occasionally reported bug whereby a message would remain on the queue
6592 longer than the retry timeout, but would be bounced if a delivery was
6593 forced. I say "may" because I never totally pinned down the problem;
6594 setting up timeout/retry tests is difficult. See also the next item.
6596 79. The ultimate address timeout was not being applied to errors that involved
6597 a combination of host plus message (for example, a timeout on a MAIL
6598 command). When an address resolved to a number of possible hosts, and they
6599 were not all tried for each delivery (e.g. because of hosts_max_try), a
6600 message could remain on the queue longer than the retry timeout.
6602 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6605 81. Fixed an obscure SMTP outgoing bug which required at least the following
6606 conditions: (a) there was another message waiting for the same server;
6607 (b) the server returned 5xx to all RCPT commands in the first message so
6608 that the message was not completed; (c) the server dropped the connection
6609 or gave a negative response to the RSET that Exim sends to abort the
6610 transaction. The observed case was a dropped connection after DATA that had
6611 been sent in pipelining mode. That is, the server had advertised PIPELINING
6612 but was not implementing it correctly. The effect of the bug was incorrect
6613 behaviour, such as trying another host, and this could lead to a crash.
6619 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6620 and daemon.c were passed as pointers to ints; they should have been
6621 pointers to socklen_t variables (which are typically unsigned ints).
6623 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6626 3. Fixed a really odd bug that affected only the testing scheme; patching a
6627 certain fixed string in the binary changed the value of another string that
6628 happened to be identical to the end of the original first string.
6630 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6631 name", it returns that address as the IP address. On some operating
6632 systems (e.g. Solaris), it also passes back the IP address string as the
6633 "host name". However, on others (e.g. Linux), it passes back an empty
6634 string. Exim wasn't checking for this, and was changing the host name to an
6635 empty string, assuming it had been canonicalized.
6637 5. Although rare, it is permitted to have more than one PTR record for a given
6638 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6639 all the names associated with an address, because they do in Solaris.
6640 However, it seems that they do not in Linux for data that comes from the
6641 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6642 I found this out when I moved to a new Linux workstation and tried to run
6643 the Exim test suite.
6645 To get round this problem I have changed the code so that it now does its
6646 own call to the DNS to look up PTR records when searching for a host name.
6647 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6648 addresses that are only in /etc/hosts are still found.
6650 This behaviour is, however, controlled by an option called host_lookup_
6651 order, which defaults to "bydns:byaddr". If people want to use the other
6652 order, or indeed, just use one or the other means of lookup, they can
6653 specify it in this variable.
6655 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6656 some operating systems, this comes back from gethostbyaddr() as an empty
6657 string, and this is what Exim used to test for. However, it seems that in
6658 other systems, "." is yielded. Exim now tests for this case too.
6660 7. The values of check_spool_space and check_log_space are now held internally
6661 as a number of kilobytes instead of an absolute number of bytes. If a
6662 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6663 kilobyte. This means that much larger values can be stored.
6665 8. Exim monitor: an attempt to get the action menu when not actually pointing
6666 at a message produces an empty menu entitled "No message selected". This
6667 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6668 no entries in it ("Shell widget menu has zero width and/or height"). So I
6669 have added a single, blank menu entry in this case.
6671 9. Added ${quote_local_part.
6673 10. MIME decoding is now applied to the contents of Subject: header lines when
6676 11. Now that a reference to $sender_host_address automatically causes a reverse
6677 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6678 host lookup before query-style lookups in lists that might use this
6679 variable. This has therefore been abolished, and the "net-" prefix is no
6680 longer necessary for query-style lookups.
6682 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6683 have been a typo for LFLAGS, so it has been changed.
6685 13. The install script calls Exim with "-C /dev/null" in order to find the
6686 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6687 to be output. However, since Exim outputs its version number before the
6688 error, it didn't break the script. It just looked ugly. I fixed this by
6689 always allowing "-C /dev/null" if the caller is root.
6691 14. Ignore overlarge ACL variable number when reading spool file - insurance
6692 against a later release with more variables having written the file.
6694 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6695 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6696 this, as well as the form without the "IPv6" on the front.
6698 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6699 OS/Makefile-Darwin file.
6701 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6702 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6704 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6705 variable. However, some OS use ints for inodes. I've added cast to long int
6706 to get rid of the compiler warning.
6708 19. I had forgotten to lock out "/../" in configuration file names when
6709 ALT_CONFIG_PREFIX was set.
6711 20. Routers used for verification do not need to specify transports. However,
6712 if such a router generated a host list, and callout was configured, Exim
6713 crashed, because it could not find a port number from the (non-existent)
6714 transport. It now assumes port 25 in this circumstance.
6716 21. Added the -t option to exigrep.
6718 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6719 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6720 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6721 them is set, LOOKUP_LSEARCH is forced.
6723 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6725 24. Added sender and host information to the "rejected by local_scan()" log
6726 line; previously there was no indication of these.
6728 25. Added .include_if_exists.
6730 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6731 Linux. It turns out that not all file systems support this. Apparently some
6732 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6733 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6734 not supported on the file descriptor, is now ignored when Exim is trying to
6735 sync a directory. This applies only to Linux.
6737 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6739 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6740 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6743 29. Added support for maildirsize files from supplied patch (modified a bit).
6745 30. The use of :fail: followed by an empty string could lead Exim to respond to
6746 sender verification failures with (e.g.):
6748 550 Verification failed for <xxx>
6749 550 Sender verify failed
6751 where the first response line was missing the '-' that indicates it is not
6752 the final line of the response.
6754 31. The loop for finding the name of the user that called Exim had a hardwired
6755 limit of 10; it now uses the value of finduser_retries, which is used for
6756 all other user lookups.
6758 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6761 33. Exim was neglecting to zero errno before one call of strtol() when
6762 expanding a string and expecting an integer value. On some systems this
6763 resulted in spurious "integer overflow" errors. Also, it was casting the
6764 result into an int without checking.
6766 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6767 did not work. The code looks as if it has *never* worked, though it appears
6768 to have been documented since at least release 1.62. I have made it work.
6770 35. The "timeout_DNS" error in retry rules, also documented since at least
6771 1.62, also never worked. As it isn't clear exactly what this means, and
6772 clearly it isn't a major issue, I have abolished the feature by treating it
6773 as "timeout", and writing a warning to the main and panic logs.
6775 36. The display of retry rules for -brt wasn't always showing the error code
6778 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6779 timeout_connect_A, timeout_connect_MX.
6781 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6782 to the empty sender.
6784 39. The daemon was not analysing the content of -oX till after it had closed
6785 stderr and disconnected from the controlling terminal. This meant that any
6786 syntax errors were only noted on the panic log, and the return code from
6787 the command was 0. By re-arranging the code a little, I've made the
6788 decoding happen first, so such errors now appear on stderr, and the return
6789 code is 1. However, the actual setting up of the sockets still happens in
6790 the disconnected process, so errors there are still only recorded on the
6793 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6794 connections (as happens on some IP stacks) was logged at start up time as
6795 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6796 it from "IPv6 and IPv4", which means that two separate sockets are being
6799 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6800 says whether AF_INET or AF_INET6 was passed as an argument.
6802 42. Exiwhat output was messed up when time zones were included in log
6805 43. Exiwhat now gives more information about the daemon's listening ports,
6806 and whether -tls-on-connect was used.
6808 44. The "port" option of the smtp transport is now expanded.
6810 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6811 silently ignored. Now an error message is written to the main and panic
6814 46. There's a new ACL modifier called "logwrite" which writes to a log file
6815 as soon as it is encountered.
6817 47. Added $local_user_uid and $local_user_gid at routing time.
6819 48. Exim crashed when trying to verify a sender address that was being
6822 49. Exim was recognizing only a space character after ".include". It now also
6823 recognizes a tab character.
6825 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6826 extracting the relevant information from the specification. The man page no
6827 longer contains scrambled data for the -d option, and I've added a section
6828 at the front about calling Exim under different names.
6830 51. Added "extra_headers" argument to the "mail" command in filter files.
6832 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6835 53. Installed eximstats 1.29.
6837 54. Added transport_filter_timeout as a generic transport option.
6839 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6840 Cc: header lines. This was required by RFC 822, but it not required by RFC
6843 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6844 incoming messages that did not have them. Now it does so only if the
6845 message originates locally, that is, if there is no associated remote host
6846 address. When Resent- header lines are present, this applies to the Resent-
6847 lines rather than the non-Resent- lines.
6849 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6850 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6852 58. Messages for configuration errors now include the name of the main
6853 configuration file - useful now that there may be more than one file in a
6854 list (.included file names were always shown).
6856 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6857 for those rare installations that do not start the daemon as root or run it
6858 setuid root. I've cut out the call to initgroups() if the daemon is not
6861 60. The Exim user and group can now be bound into the binary as text strings
6862 that are looked up at the start of Exim's processing.
6864 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6866 62. Added $mailstore_basename variable.
6868 63. Installed patch to sieve.c from Michael Haardt.
6870 64. When Exim failed to open the panic log after failing to open the main log,
6871 the original message it was trying to log was written to stderr and debug
6872 output, but if they were not available (the usual case in production), it
6873 was lost. Now it is written to syslog before the two lines that record the
6874 failures to open the logs.
6876 65. Users' Exim filters run in subprocesses under the user's uid. It is
6877 possible for a "deliver" command or an alias in a "personal" command to
6878 provoke an address rewrite. If logging of address rewriting is configured,
6879 this fails because the process is not running as root or exim. There may be
6880 a better way of dealing with this, but for the moment (because 4.30 needs
6881 to be released), I have disabled address rewrite logging when running a
6882 filter in a non-root, non-exim process.
6888 1. The buildconfig auxiliary program wasn't quoting the value set for
6889 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6890 not defined. This bug was masked in 4.22 by the effect that was fixed in
6893 2. Some messages that were rejected after a message id was allocated were
6894 shown as "incomplete" by exigrep. It no longer does this for messages that
6895 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6897 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6898 have allow_domain_literals set, the ID did not get logged in the <= line.
6899 Domain literals are now always recognized in Message-ID: header lines.
6901 4. The first argument for a ${extract expansion item is the key name or field
6902 number. Leading and trailing spaces in this item were not being ignored,
6903 causing some misleading effects.
6905 5. When deliver_drop_privilege was set, single queue runner processes started
6906 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6907 same command in the process it spins off) were not dropping privilege.
6909 6. When the daemon running as "exim" started a queue runner, it always
6910 re-executed Exim in the spun-off process. This is a waste of effort when
6911 deliver_drop_privilege is set. The new process now just calls the
6912 queue-runner function directly.
6918 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6921 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6922 ignored. Though the use of -oP was forcing the writing of a pid file, it
6923 was always written to the default place.
6925 3. If the message "no IP address found for host xxxx" is generated during
6926 incoming verification, it is now followed by identification of the incoming
6927 connection (so you can more easily find what provoked it).
6929 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6931 5. Added some features to "harden" Exim a bit more against certain attacks:
6933 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6934 be put in Local/Makefile. This is like the never_users runtime option,
6935 but it cannot be overridden. The default setting is "root".
6937 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6938 prefix string with which any file named in a -C command line option
6941 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6942 is retained for -C and -D only if the caller of Exim is root. Without
6943 it, the exim user may also use -C and -D and retain privilege.
6945 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6946 command line option is disabled.
6948 6. Macro names set by the -D option must start with an upper case letter, just
6949 like macro names defined in the configuration file.
6951 7. Added "dereference=" facility to LDAP.
6953 8. Two instances of the typo "uknown" in the source files are fixed.
6955 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6956 the Configure-Makefile script screwed up while processing it.
6958 10. Incorporated PCRE 4.4.
6960 11. The SMTP synchronization check was not operating right at the start of an
6961 SMTP session. For example, it could not catch a HELO sent before the client
6962 waited for the greeting. There is now a check for outstanding input at the
6963 point when the greeting is written. Because of the duplex, asynchronous
6964 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6965 way, but not yet received, when the check is performed.
6967 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6968 on TCP/IP sockets, because this apparently causes some broken clients to
6971 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6972 unchanged) from the Cygwin maintainer.
6974 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6975 was supported instead of testing for mbx. Effectively a typo.
6977 15. The spa authenticator server code was not checking that the input it
6978 received was valid base64.
6980 16. The debug output line for the "set" modifier in ACLs was not showing the
6981 name of the variable that was being set.
6983 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6985 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6986 lookup on its own. Something else was needed to trigger the lookup. For
6987 example, a match in host_lookup or the need for a host name in a host list.
6988 Now, if $sender_host_name is referenced and the host name has not yet been
6989 looked up, a lookup is performed. If the lookup fails, the variable remains
6990 empty, and $host_lookup_failed is set to "1".
6992 19. Added "eqi" as a case-independent comparison operator.
6994 20. The saslauthd authentication condition could segfault if neither service
6995 nor realm was specified.
6997 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6998 error message that was logged was misleading, and incoming SMTP
6999 connections were dropped. The message is now more accurate, and temporary
7000 errors are given to SMTP connections.
7002 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
7003 (except RSET) with a 421 error, until QUIT is received. However, it was
7004 failing to send a response to QUIT.
7006 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
7007 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
7008 ACL is now run after the helo_try_verify_hosts code.
7010 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
7011 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
7012 case-independent, so other case variants are also recognized). Apparently
7013 some systems use these upper case variants.
7015 25. If more than two messages were waiting for the same host, and a transport
7016 filter was specified for the transport, Exim sent two messages over the
7017 same TCP/IP connection, and then failed with "socket operation on non-
7018 socket" when it tried to send the third.
7020 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
7022 27. The extern definition of crypt16() in expand.c was not being excluded when
7023 the OS had its own crypt16() function.
7025 28. Added bounce_return_body as a new option, and bounce_return_size_limit
7026 as a preferred synonym for return_size_limit, both as an option and as an
7029 29. Added LIBS=-liconv to OS/Makefile-OSF1.
7031 30. Changed the default configuration ACL to relax the local part checking rule
7032 for addresses that are not in any local domains. For these addresses,
7033 slashes and pipe symbols are allowed within local parts, but the sequence
7034 /../ is explicitly forbidden.
7036 31. SPA server authentication was not clearing the challenge buffer before
7039 32. log_message in a "warn" ACL statement was writing to the reject log as
7040 well as to the main log, which contradicts the documentation and doesn't
7041 seem right (because no rejection is happening). So I have stopped it.
7043 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
7044 However, I am unable to do any testing of this.
7046 34. Fixed an infelicity in the appendfile transport. When checking directories
7047 for a mailbox, to see if any needed to be created, it was accidentally
7048 using path names with one or more superfluous leading slashes; tracing
7049 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
7051 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
7052 contents are added to the log line that is written for every discarded
7053 recipient. (Previously a log_message setting was ignored.)
7055 36. The ${quote: operator now quotes the string if it is empty.
7057 37. The install script runs exim in order to find its version number. If for
7058 some reason other than non-existence or emptiness, which it checks, it
7059 could not run './exim', it was installing it with an empty version number,
7060 i.e. as "exim-". This error state is now caught, and the installation is
7063 38. An argument was missing from the function that creates an error message
7064 when Exim fails to connect to the socket for saslauthd authentication.
7065 This could cause Exim to crash, or give a corrupted message.
7067 39. Added isip, isip4, and isip6 to ${if conditions.
7069 40. The ACL variables $acl_xx are now saved with the message, and can be
7070 accessed later in routers, transports, and filters.
7072 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
7073 strings in the file are not string-expanded.
7075 42. If a MAIL command specified a SIZE value that was too large to fit into an
7076 int variable, the check against message_size_limit failed. Such values are
7077 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
7078 day this will have to be increased, but I don't think I want to be around
7079 when emails are that large.
7086 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
7087 iconv() is not standard in FreeBSD.
7089 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
7090 IPv6 enabled. The observed symptom was a segmentation fault on return from
7091 the function os_common_find_running_interfaces() in src/os.c.
7093 3. In the check_special_case() function in daemon.c I had used "errno" as an
7094 argument name, which causes warnings on some systems. This was basically a
7095 typo, since it was named "eno" in the comments!
7097 4. The code that waits for the clock to tick (at a resolution of some fraction
7098 of a second) so as to ensure message-id uniqueness was always waiting for
7099 at least one whole tick, when it could have waited for less. [This is
7100 almost certainly not relevant at current processor speeds, where it is
7101 unlikely to ever wait at all. But we try to future-proof.]
7103 5. The function that sleeps for a time interval that includes fractions of a
7104 second contained a race. It did not block SIGALRM between setting the
7105 timer, and suspending (a couple of lines later). If the interval was short
7106 and the sigsuspend() was delayed until after it had expired, the suspension
7107 never ended. On busy systems this could lead to processes getting stuck for
7110 6. Some uncommon configurations may cause a lookup to happen in a queue runner
7111 process, before it forks any delivery processes. The open lookup caching
7112 mechanism meant that the open file or database connection was passed into
7113 the delivery process. The problem was that delivery processes always tidy
7114 up cached lookup data. This could cause a problem for the next delivery
7115 process started by the queue runner, because the external queue runner
7116 process does not know about the closure. So the next delivery process
7117 still has data in the lookup cache. In the case of a file lookup, there was
7118 no problem because closing a file descriptor in a subprocess doesn't affect
7119 the parent. However, if the lookup was caching a connection to a database,
7120 the connection was closed, and the second delivery process was likely to
7121 see errors such as "PGSQL: query failed: server closed the connection
7122 unexpectedly". The problem has been fixed by closing all cached lookups
7123 in a queue runner before running a delivery process.
7125 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
7126 seem to have the "const" qualifier which it has on other OS. I've
7129 8. Change 4.21/2 was too strict. It is only if there are two authenticators
7130 *of the same type* (client or server) with the same public name that an
7131 error should be diagnosed.
7133 9. When Exim looked up a host name for an IP address, but failed to find the
7134 original IP address when looking up the host name (a safety check), it
7135 output the message "<ip address> does not match any IP for NULL", which was
7136 confusing, to say the least. The bug was that the host name should have
7137 appeared instead of "NULL".
7139 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
7140 user that is built into the binary, and the -C or -D options is used, root
7141 privilege is dropped before the configuration file is read. In addition,
7142 logging is switched to stderr instead of the normal log files. If the
7143 configuration then re-defines the Exim user, the unprivileged environment
7144 is probably not what is expected, so Exim logs a panic warning message (but
7147 However, if deliver_drop_privilege is set, the unprivileged state may well
7148 be exactly what is intended, so the warning has been cut out in that case,
7149 and Exim is allowed to try to write to its normal log files.
7155 1. smtp_return_error_details was not giving details for temporary sender
7156 or receiver verification errors.
7158 2. Diagnose a configuration error if two authenticators have the same public
7161 3. Exim used not to create the message log file for a message until the first
7162 delivery attempt. This could be confusing when incoming messages were held
7163 for policy or load reasons. The message log file is now created at the time
7164 the message is received, and an initial "Received" line is written to it.
7166 4. The automatically generated man page for command line options had a minor
7167 bug that caused no ill effects; however, a more serious problem was that
7168 the procedure for building the man page automatically didn't always
7169 operate. Consequently, release 4.20 contains an out-of-date version. This
7170 shouldn't happen again.
7172 5. When building Exim with embedded Perl support, the script that builds the
7173 Makefile was calling 'perl' to find its compile-time parameters, ignoring
7174 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
7176 6. The freeze_tell option was not being used for messages that were frozen on
7177 arrival, either by an ACL or by local_scan().
7179 7. Added the smtp_incomplete_transaction log selector.
7181 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
7182 was accepting AUTH without a new EHLO.
7184 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
7185 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
7188 10. Logging of TCP/IP connections (when configured) now happens in the main
7189 daemon process instead of the child process, so that the TCP/IP connection
7190 count is more accurate (but it can never be perfect).
7192 11. The use of "drop" in a nested ACL was not being handled correctly in the
7193 outer ACL. Now, if condition failure induced by the nested "drop" causes
7194 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
7195 or "require"), the connection is dropped.
7197 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
7198 that yield "discard" can now be used with an "accept" or a "discard" verb,
7199 but an error is generated for any others (because I can't see a useful way
7200 to define what should happen).
7202 13. When an ACL is read dynamically from a file (or anywhere else), the lines
7203 are now processed in the same way as lines in the Exim configuration file.
7204 In particular, continuation lines are supported.
7206 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
7208 15. Added -ti meaning -t -i.
7210 16. Check for letters, digits, hyphens, and dots in the names of dnslist
7211 domains, and warn by logging if others are found.
7213 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
7214 returned from GIFCONF when Exim is trying to find the list of interfaces on
7215 a host. The code in os.c has been modified to copy each ifreq to an aligned
7216 structure in all cases.
7218 Also, in some cases, the returned ifreq's were being copied to a 'struct
7219 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
7220 means the last couple of bytes of an IPv6 address could be chopped if the
7221 ifreq contained only a normal sockaddr (14 bytes storage).
7223 18. Named domain lists were not supported in the hosts_treat_as_local option.
7224 An entry such as +xxxx was not recognized, and was treated as a literal
7227 19. Ensure that header lines added by a DATA ACL are included in the reject log
7228 if the ACL subsequently rejects the message.
7230 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
7231 MD5 (which is deprecated).
7233 21. When testing a filter file using -bf, Exim was writing a message when it
7234 took the sender from a "From " line in the message, but it was not doing so
7235 when it took $return_path from a Return-Path: header line. It now does.
7237 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
7238 with a valid header line field name (a series of printing characters
7239 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
7241 23. Changed "disc" in the source to "disk" to conform to the documentation and
7242 the book and for uniformity.
7244 24. Ignore Sendmail's -Ooption=value command line item.
7246 25. When execve() failed while trying to run a command in a pipe transport,
7247 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
7248 could be confused with a return value of 69 from the command itself. This
7249 has been changed to 127, the value the shell returns if it is asked to run
7250 a non-existent command. The wording for the related log line suggests a
7251 non-existent command as the problem.
7253 26. If received_header_text expands to an empty string, do not add a Received:
7254 header line to the message. (Well, it adds a token one on the spool, but
7255 marks it "old" so that it doesn't get used or transmitted.)
7257 27. Installed eximstats 1.28 (addition of -nt option).
7259 28. There was no check for failure on the call to getsockname() in the daemon
7260 code. This can fail if there is a shortage of resources on the system, with
7261 ENOMEM, for example. A temporary error is now given on failure.
7263 29. Contrary to the C standard, it seems that in some environments, the
7264 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
7265 program. Exim now does this explicitly; it affects the formatting of
7266 timestamps using strftime().
7268 30. If exiqsumm was given junk data, it threw up some uninitialized variable
7269 complaints. I've now initialized all the variables, to avoid this.
7271 32. Header lines added by a system filter were not being "seen" during
7272 transport-time rewrites.
7274 33. The info_callback() function passed to OpenSSL is set up with type void
7275 (*)(SSL *, int, int), as described somewhere. However, when calling the
7276 function (actually a macro) that sets it up, the type void(*)() is
7277 expected. I've put in a cast to prevent warnings from picky compilers.
7279 34. If a DNS black list lookup found a CNAME record, but there were no A
7280 records associated with the domain it pointed at, Exim crashed.
7282 35. If a DNS black list lookup returned more than one A record, Exim ignored
7283 all but the first. It now scans all returned addresses if a particular IP
7284 value is being sought. In this situation, the contents of the
7285 $dnslist_value variable are a list of all the addresses, separated by a
7288 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
7289 to accept a host name and all its aliases if the forward lookup for any of
7290 them yielded the IP address of the incoming connection. Now it accepts only
7291 those names whose forward lookup yields the correct IP address. Any other
7292 names are discarded. This closes a loophole whereby a rogue DNS
7293 administrator could create reverse DNS records to break through a
7294 wildcarded host restriction in an ACL.
7296 37. If a user filter or a system filter that ran in a subprocess used any of
7297 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
7298 the wrong values were passed to the pipe command ($thisaddress had the
7299 value of $0, $0 had the value of $1, etc). This bug was introduced by
7300 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
7302 38. Improved the line breaking for long SMTP error messages from ACLs.
7303 Previously, if there was no break point between 40 and 75 characters, Exim
7304 left the rest of the message alone. Two changes have been made: (a) I've
7305 reduced the minimum length to 35 characters; (b) if it can't find a break
7306 point between 35 and 75 characters, it looks ahead and uses the first one
7307 that it finds. This may give the occasional overlong line, but at least the
7308 remaining text gets split now.
7310 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
7311 file descriptors might be low, and that setting 1000 would always raise it.
7312 It turns out that in some environments, the limit is already over 1000 and
7313 that lowering it causes trouble. So now Exim takes care not to decrease it.
7315 40. When delivering a message, the value of $return_path is set to $sender_
7316 address at the start of routing (routers may change the value). By an
7317 oversight, this default was not being set up when an address was tested by
7318 -bt or -bv, which affected the outcome if any router or filter referred to
7321 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
7322 log only when "message" or "log_message" are set. However, if one of the
7323 conditions was an address verification, or a call to a nested ACL, the
7324 messages generated by the underlying test were being passed through. This
7325 no longer happens. The underlying message is available in $acl_verify_
7326 message for both "message" and "log_message" expansions, so it can be
7327 passed through if needed.
7329 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
7330 new expansion $bh_ to give the encoded byte string without charset
7331 translation. Translation happens only if iconv() is available; HAVE_ICONV
7332 indicates this at build time. HEADERS_CHARSET gives the charset to
7333 translate to; headers_charset can change it in the configuration, and
7334 "headers charset" can change it in an individual filter file.
7336 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
7337 that creates RFC 2047 encoded "words" labels them as that charset instead
7338 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
7339 expansion operator; (ii) when Exim creates a From: line for a local
7340 message; (iii) when a header line is rewritten to include a "phrase" part.
7342 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
7343 buggy, causing it to skip the first lines of messages whose message ID
7344 ended in 'D'. This would not have bitten before Exim release 4.14, because
7345 message IDs were unlikely to end in 'D' before then. The effect was to have
7346 incorrect size information for certain domains.
7348 45. #include "config.h" was missing at the start of the crypt16.c module. This
7349 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
7352 46. If there was a timeout during a "random" callout check, Exim treated it as
7353 a failure of the random address, and carried on sending RSET and the real
7354 address. If the delay was just some slowness somewhere, the response to the
7355 original RCPT would be taken as a response to RSET and so on, causing
7356 mayhem of various kinds.
7358 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
7359 when I implemented it. It didn't allow for the fact that some option values
7360 may legitimately be negative (e.g. size_addition), and it didn't even do
7361 the right test for positive values.
7363 48. Domain names in DNS records are case-independent. Exim always looks them up
7364 in lower case. Some resolvers return domain names in exactly the case they
7365 appear in the zone file, that is, they may contain uppercase letters. Not
7366 all resolvers do this - some return always lower case. Exim was treating a
7367 change of case by a resolver as a change of domain, similar to a widening
7368 of a domain abbreviation. This triggered its re-routing code and so it was
7369 trying to route what was effectively the same domain again. This normally
7370 caused routing to fail (because the router wouldn't handle the domain
7371 twice). Now Exim checks for this case specially, and just changes the
7372 casing of the domain that it ultimately uses when it transmits the message
7375 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
7378 50. If a filter generated a file delivery with a non-absolute name (possible if
7379 no home directory exists for the router), the forbid_file option was not
7382 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
7383 the existing equality matching.
7385 52. Exim was using ints instead of ino_t variables in some places where it was
7386 dealing with inode numbers.
7388 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
7389 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
7390 called TMPDIR, and if it finds it is different, it changes its value.
7392 54. The smtp_printf() function is now made available to local_scan() so
7393 additional output lines can be written before returning. There is also an
7394 smtp_fflush() function to enable the detection of a dropped connection.
7395 The variables smtp_input and smtp_batched_input are exported to
7398 55. Changed the default runtime configuration: the message "Unknown user"
7399 has been removed from the ACL, and instead placed on the localuser router,
7400 using the cannot_route_message feature. This means that any verification
7401 failures that generate their own messages won't get overridden. Similarly,
7402 the "Unrouteable address" message that was in the ACL for unverifiable
7403 relay addresses has also been removed.
7405 56. Added hosts_avoid_esmtp to the smtp transport.
7407 57. The exicyclog script was not checking for the esoteric option
7408 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
7409 will work only if exicyclog is run under the appropriate euid.
7411 58. Following a discussion on the list, the rules by which Exim recognises line
7412 endings on incoming messages have been changed. The -dropcr and drop_cr
7413 options are now no-ops, retained only for backwards compatibility. The
7414 following line terminators are recognized: LF CRLF CR. However, special
7415 processing applies to CR:
7417 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
7418 nor a local message in the state where . is a terminator.
7420 (ii) If a bare CR is encountered in a header line, an extra space is added
7421 after the line terminator so as not to end the header. The reasoning
7422 behind this is that bare CRs in header lines are most likely either
7423 to be mistakes, or people trying to play silly games.
7425 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
7426 was being incorrectly given as 18 bytes larger than it should have been.
7427 This is a VOB (very old bug).
7429 60. This may never have affected anything current, but just in case it has:
7430 When the local host is found other than at the start of a list of hosts,
7431 the local host, those with the same MX, and any that follow, are discarded.
7432 When the list in question was part of a longer list of hosts, the following
7433 hosts (not currently being processed) were also being discarded. This no
7434 longer happens. I'm not sure if this situation could ever has previously
7437 61. Added the "/MX" feature to lists of hosts in the manualroute and query
7440 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
7441 header. This is something that is recommended in a new Internet Draft, and
7442 is something that is documented as being done by Sendmail. There are two
7443 possible values. For messages generated by the autoreply transport, Exim
7446 Auto-Submitted: auto-replied
7448 whereas for all other generated messages (e.g. bounces) it adds
7450 Auto-Submitted: auto-generated
7452 63. The "personal" condition in filters now includes a test for the
7453 Auto-Submitted: header. If it contains the string "auto-" the message it
7454 not considered personal.
7456 64. Added rcpt_include_affixes as a generic transport option.
7458 65. Added queue_only_override (default true).
7460 66. Added the syslog_duplication option.
7462 67. If what should have been the first header line of a message consisted of
7463 a space followed by a colon, Exim was mis-interpreting it as a header line.
7464 It isn't of course - it is syntactically invalid and should therefore be
7465 treated as the start of the message body. The misbehaviour could have
7466 caused a number of strange effects, including loss of data in subsequent
7467 header lines, and spool format errors.
7469 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
7470 client host had authenticated. This control can now be exercised by an ACL
7471 for more flexibility.
7473 69. By default, callouts do not happen when testing with -bh. There is now a
7474 variant, -bhc, which does actually run the callout code, including
7475 consulting and updating the callout cache.
7477 70. Added support for saslauthd authentication, courtesy of Alexander
7480 71. If statvfs() failed on the spool or log directories while checking their
7481 size for availability, Exim confusingly gave the error "space shortage".
7482 Furthermore, in debugging mode it crashed with a floating point exception.
7483 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7484 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7485 serious problem, Exim now writes to the main and panic logs when this
7486 happens, with details of the failure. It then refuses to accept the
7487 incoming message, giving the message "spool directory problem" or "log
7488 directory problem" with a 421 code for SMTP messages.
7490 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7491 0, 1, and 2 exist, because some OS complain for execs without them (see
7492 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7493 descriptors. However, the code omitted to check that the open succeeded,
7494 causing mysterious errors if for some reason the permissions on /dev/null
7495 got screwed. Now Exim writes a message to the main and panic logs, and
7496 bombs out if it can't open /dev/null.
7498 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7499 interact so that it is all more flexible. It is supposed to remain
7500 backwards compatible. Also added extra_local_interfaces.
7502 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7503 to bomb out with an assertion failure - to the client this appears as a
7504 connection drop. This problem occurs in the part of the code that was taken
7505 from the Samba project. Fortunately, the assertion is in a very simple
7506 function, so I have fixed this by reproducing the function inline in the
7507 one place where it is called, and arranging for authentication to fail
7508 instead of killing the process with assert().
7510 75. The SPA client code was not working when the server requested OEM rather
7511 than Unicode encoding.
7513 76. Added code to make require_files with a specific uid setting more usable in
7514 the case where statting the file as root fails - usually a non-root-mounted
7515 NFS file system. When this happens and the failure is EACCES, Exim now
7516 forks a subprocess and does the per-uid checking as the relevant uid.
7518 77. Added process_log_path.
7520 78. If log_file_path was not explicitly set, a setting of check_log_space or
7521 check_log_inodes was ignored.
7523 79. If a space check for the spool or log partitions fails, the incident is now
7524 logged. Of course, in the latter case the data may get lost...
7526 80. Added the %p formatting code to string_format() so that it can be used to
7527 print addresses in debug_print(). Adjusted all the address printing in the
7528 debugging in store.c to use %p rather than %d.
7530 81. There was a concern that a line of code in smtp_in.c could overflow a
7531 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7532 initially expressed, the concern was not well-founded, because trailing
7533 spaces are removed early. However, if the trailing spaces were followed by
7534 a NULL, they did not get removed, so the overflow was possible. Two fixes
7537 (a) I re-wrote the offending code in a cleaner fashion.
7538 (b) If an incoming SMTP command contains a NULL character, it is rejected
7541 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7542 runs initgroups(), so that if the Exim user is in any additional groups,
7543 they will be used during message reception.
7549 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / heiko / exim.git / blob